Do not remove my spam

Modern internet is broken and no longer built for human beings. By 2026 the internet is filled with robots. One type of robots generates content and injects advertisements, another type ingests that content and emulates human engagement metrics. Modern social media platforms treat spam as something evil that must be automatically removed. But the irony is that commercial spam was never removed. It became the dominant form of communication on the internet.

Most apps and websites are not built to make people happy or help them discover meaningful things. They are built to maximize engagement, retention, and monetizable behavior.

This is why social media feeds are filled with:

• political outrage
• influencer agenda
• AI slop
• engagement bait
• endless “interesting facts” that do not affect your life
• productivity roleplay
• pseudo-educational garbage
• and emotionally manipulative content optimized for metrics

The platforms decide what communication formats are considered legitimate.

You are encouraged to:

• consume,
• scroll,
• react,
• subscribe,
• watch,
• engage.

But unsolicited communication increasingly becomes taboo.

Some people no longer see anything strange in apps like Tinder, which are optimized primarily for retention and monetization rather than human outcomes. They spend years inside systems designed to keep them engaged while slowly starting to perceive approaching strangers in public as abnormal or suspicious behavior.

The platforms define what forms of interaction feel socially acceptable.

The same thing happened to publishing.

The old internet was built around uncontrolled discovery:

• personal blogs,
• weird forums,
• RSS feeds,
• forum signatures,
• random hyperlinks,
• blogrolls,
• obscure comments,
• niche mailing lists.

You could discover genuinely interesting people accidentally.

Now discovery is increasingly permission-based.

The platform decides:

• who can reach people,
• what links are acceptable,
• what visibility is allowed,
• what promotion is “organic”,
• and what communication is classified as spam.

Tinder-like apps make approaching strangers in public more frowned upon. Social media automatically moderates content for users, and sells the idea that spam is bad.

Don’t buy it.

Spam is totally fine

Let’s talk honestly about what spam actually is.

When I check my email inbox, most messages are generated automatically. Most are irrelevant. Many are so bad that the mail provider immediately moves them into the junk folder. Some messages survive the spam filters, but when I open them, I see call-to-action requests without any insightful information, so I click the button and use these letters to train my spam filter.

Sometimes I see brilliant messages that have been sent via automation. I noticed they never look like “Do you want to get known about it? Contact me if you are interesting!”, instead, they include all the information, they are insightful, and I learn something new when I read such messages.

The sender’s intention to sell me something does not matter to me. If I learned something valuable, the communication already succeeded. Sometimes I am interested in their service and I click links, but mostly I don’t.

Social media tries to marginalize irrelevant messages they have not been paid for. As a result, they hide uncut gems from us.

The fight against spam

I have an account on X, and every time I try to open the feed page, I see totally irrelevant content - random screenshots with code and “beautiful” interfaces, “insights” about AI, “interesting” facts, influencer trash talk. It’s just because Twitter thinks I’m a developer, so AI garbage may be interesting to me.

However I don’t see in my feed content from random persons with short and meaningful blog posts.

That’s because the Twitter algorithms show commercial spam above others. People will see your posts in their feed only if you pay for promotion, pro plan, etc.

I’ve tried to find ways to read independent writers, and I’ve tried to publish my own writings. I mostly ended up with deletions of my links and bans.

For example, Reddit as a platform sometimes automatically deletes posts you send. It deletes your posts with no reason and randomly. You can ask the moderator of the “subreddit” where you posted to recover your post, and sometimes they do. But Reddit did not ask their opinion and bans your post automatically prematurely.

That does not mean Reddit has no spam. There are many well-known strategies for promoting anything on Reddit, for example, people create many accounts, fake their activity, create a thread that looks like a question, and then send their spam from other accounts. Reddit pretends it looks legit, because well-faked user activity brings money to Reddit.

As you may notice, the problem with the status quo is that only non-commercial spam can be blocked. Any spam that efficiently makes more money than it spends will not go away.

The fight against spam is mostly theater. The platforms get rid of independent non-commercial publishers. End users only see commercial spam, like influencer posts that pretend to be insightful but actually sell you yet another service they paid for.

The platforms make more money and are happy, users waste their lives seeing ads on the screen.

Missed opportunities

Recently I’ve open-sourced the NPM package to automate the translation of i18n locale files.

I tried to promote my package on GitHub. I opened the global GitHub issues list with all open issues on GitHub, found issues related to i18n, and left comments there suggesting to use transly and explaining how localization positively affects SEO results and helps reach a wide audience.

I actually tried to solve relevant problems for relevant people.

I’ve got a few stars on GitHub, but most people just ignored my spam.

I tried to create issues with the same content in a few boilerplate repositories. Most of the issues were closed with no comments or comments like this:

It seems to be a spam, the exactly the same message into multiple repositories.

It seems most people on GitHub are highly resistant to spam. They don’t engage with my proposal substantively just because it was automated.

It no longer matters:

• whether the proposal is useful,
• whether the tool solves real problem,
• whether users may benefit from it,
• whether the recommendation is relevant.

As a user of the boilerplate for your landing site or SaaS, you will lose some SEO benefits and your project will be artificially inaccessible to a wide audience just because the maintainer who created your boilerplate has stale knowledge and does not seriously consider proposals that have been sent to a few repositories.

The consequence of avoiding any marginalia is missed opportunities.

Do not remove a spam

Well okay, if modern internet mostly shows commercial spam and engagement farming, what should we actually do?

My answer is: don’t block spam. Stop demanding perfectly sanitized platforms.

Moreover, if you want to find worthwhile content, you should look for more spam.

The less moderation and auto filters a platform has, the greater the chance to find something interesting.

Look for marginal platforms where spam is not blocked well yet. For example, instead of Twitter, try Bluesky or Mastodon, they are flooded with spam, so you can just filter it by interesting tags, ban a few accounts that appear frequently with garbage, and sometimes you will find interesting content.

Consider spam as a raw data in your Clickhouse DB. You may transform and filter it anytime and get valuable insights if you have enough data to process.

You can manually ignore annoying accounts. You can block obvious garbage yourself. Human beings were perfectly capable of ignoring irrelevant things long before machine learning moderation existed.

Don’t lock yourself in Facebook or dating apps. Be wild, be out of the mainstream to find interesting things ahead of others.

When every interaction must be platform-approved, the internet stops being a network and becomes television again.