You took the address from wherever it appeared first
Mars market mirrors
marshjhtog245vzjzcicnmv2ci6yljibvdm4pngq5kmkfvcutppboxad.onion
marsiujka6lrsaqpnxiwvknthhzsrlmq77mnl2fi62guc4lwxif65syd.onion
marsmtbwtxkohhpu34m4jkwcntian7n257wsex5tbkmsjdjrsz6me3yd.onion
Printed as supplied, in no order. Nothing here is watched or timed, so an address that opens is not proof of anything. More about the set
The old address stopped opening. The reader searched for a mars market working link, opened the first page that looked confident, and used what it gave them. Later something went wrong and the story becomes a story about scammers.
Scammers were involved. They did not cause this. The decision to treat an unranked, unsourced string as authoritative was made by the reader, in about four seconds, under mild time pressure.
- Responsible
- You. A source was chosen, and choosing it was the whole act.
- Usually blamed on
- Phishers, criminals, and the general badness of the internet. All present, none of them responsible for the selection.
- What follows
- Sourcing is a decision with a quality, and it can be made better. Being angry at phishers cannot be made better.
What a search rank actually measures
It measures how well a page satisfies a ranking system. That system has no way to check whether a string of base32 characters points at the service its page claims. It reads text. A page can be careful, well linked and completely wrong, and nothing about the ranking notices.
This is worth sitting with. A page that spends real effort on being findable and a page that spends real effort on being correct look identical from the results list. The effort is going to different places and the display cannot tell them apart.
The four second problem
Nobody deliberates. The old link fails, the search happens, the first plausible page opens, and the address is in the clipboard before any evaluation takes place. The evaluation, if it happens at all, happens afterwards and consists of noticing that the page looked fine.
That sequence is the actual failure. Not the bad address. The bad address is downstream of a sourcing process that took four seconds and had no criteria in it.
Criteria that cost almost nothing
| Question to ask the page | What the answer is worth |
|---|---|
| Does the page name where it got the address | A page that cannot say is guessing too |
| Does the page publish a full set or one string | One string with no siblings is a thin claim |
| Does the page claim the address is up right now | That claim cannot be made, and making it is a tell |
| Does the page agree with a second, unrelated source | Two independent sources is not proof, but it is not nothing |
None of these is a guarantee. The point is not certainty. The point is that four seconds and no criteria can be replaced by forty seconds and four criteria, and that is the entire available improvement.
What follows from getting it right
The reader who blames scammers changes nothing about their behaviour, because scammers are a weather condition. The reader who accepts that sourcing was the decision starts keeping a set of addresses in advance, from more than one place, checked when calm. That reader is not making this mistake again.
This site publishes its addresses in one place and says plainly where the claim ends. It cannot tell you an address is safe. It can tell you it is published, and that it is printed here exactly as it was supplied.
Asked often enough to answer
Does a padlock or a certificate help me here?
No. Onion services carry their own transport encryption and a browser padlock says nothing about who runs the service. A hostile copy of a market looks exactly as secure as the real one.
Is checking the first few characters enough?
No. A prefix can be generated deliberately by anybody willing to spend the computing time, and prefixes are cheap. Every character in the address has to match.