Ome.tv and others: the claims, and which ones survive
Roughly the same four claims come from every replacement. Below, in the order they reach you.
«Thousands online right now»
- The figure counts tabs, not people mid-press.
- It includes everybody already talking.
- It includes everybody who left the tab open at lunchtime.
This counts open tabs. Somebody mid-conversation is in the figure; so is somebody who opened the page during lunch and forgot about it. On a service where the average call runs two minutes, the proportion of that number actually pressing next at any instant is small and nobody publishes it.
Which makes the headline number close to meaningless rather than dishonest. It is a real count of a thing you do not care about.
Nobody has ever published the figure you would want, which is presses per second, and the reason is that it would look small on every service in the world.
«Better matching»
Almost always a filter, and filters do work — they narrow the pool exactly as promised. The part left off the page is the price of narrowing, which is waiting, and it does not scale politely: a matching pair has to press together where formerly any pair sufficed.
Worth trying once the free version has had a fair week. Not worth buying on day one, when what you actually need is the right hour.
And the arithmetic gets worse the narrower you go. A country filter roughly halves a pool; a country filter plus a gender filter can take it down by a factor nobody advertises, and the waiting grows to match.
«Safer, with moderation»
Two layers everywhere, on every service, paid or free. The automatic one examines frames rather than conversations; its mistakes cluster in the first two seconds, while a camera is still being shifted around. The human layer works through a list days afterwards.
There is no third layer anywhere, and none of it can reach into the minute and a half you happen to be inside. A claim implying otherwise describes a wish.
«Free, with premium options»
The one claim that is straightforwardly true, and the structure underneath it is worth knowing: the free side makes money when a visitor leaves for the paid side, which is why nothing on the free side counts your minutes or builds up over visits. Nobody would earn anything from building that.
So what to do with an evening
- Check the text-only question on three services. Ten seconds each.
- Open all three in the same half hour, at a late local hour somewhere you would like to reach.
- Give each ten minutes, and time the waits rather than judging the conversations.
- Keep whichever connected fastest, and redo this in six months, because it will be a different one.
Half an evening of work, and it outperforms every published ranking, since no ranking knows which hour you are awake in and the hour is the variable that decides.