Where this is going

Plainly, and including the parts that do not work yet. If we are asking you to shape this, you need to be able to see the shape.

This is an early version and says so everywhere. What follows is what we think comes next, segment by segment, with an honest label on each: working means you can use it today, decided means it is settled but not built, open means we genuinely do not know and your answer would change it.

None of this order is fixed. We have put these in the sequence we think most people need them, and we have had that wrong before. If enough people tell us something further down the list is what they actually came for, it moves up. That is not politeness — it is the only way the order gets to be right.

The segments

Saying what is happening to you

Working · and the least finished of the working things

Your Health Story takes it in whatever shape it arrives and hands it back ordered, in your own vocabulary. It never leaves the device you are on.

Next: it asks the same question of everybody in the same voice. It should not. A frightened child, a guarded man who does not want to be here, and a grandmother who will not be hurried need the same question asked three different ways. That is designed and not built.

Getting more out of a short appointment

Working

Preparing for the Visit reduces the whole worry to a single sheet you can hold, ordered so the most important thing is said while there is still time to say it.

Next: what to do with what you were told afterwards — the half of the appointment that falls apart in the car park.

Seeing how two conditions connect

Working

Where conditions overlap draws the part in the middle that neither specialist owns, so you can point at it and ask who is watching it.

Next: more than two. Most complicated situations are three or four things at once, and the middle gets more interesting, not less.

Knowing how good the evidence actually is

Decided · partly built

Every claim carries a grade, A through F, plus U for “the evidence genuinely does not resolve this”. The scale is settled and the reasoning behind it is written down.

Next: the grades have to travel with the claims across the whole site, not sit in one explanation. And a letter is our shorthand, not yours — see the open questions below.

Telling us we got it wrong

Working, but only halfway

The note control on every page holds what you write on your own machine. Nothing travels, and we hold nothing at all without your explicit say-so.

Next, and it is the one that matters: somewhere for it to go, and a human answer to every single one, promptly and in their own words. That needs a server, which is days away rather than months.

Doing your own research, and having it count

Method working · the collecting part not built

How to record what you noticed so it survives a year and an argument is on the site now and needs nothing from us. Alone it is one person's story; gathered, the same story told four hundred times is evidence that has never been collected.

Next: the place to put it, and the ability to see what other people put there. Open to anyone — patients, carers, students, researchers, and people from trades with nothing to do with medicine, who are often the ones who see it.

Arguing about it in the open

Decided · needs the server

A room where the answering happens in public, where anyone who thinks we misread the evidence can say so and have it out, and where the half-finished work stays visible instead of being tidied away until it looks certain.

The harder half — carrying knowledge across fields

Ongoing, and mostly invisible from here

An answer sitting finished in one discipline, and a question still open in another, with no door between them. Finding those is slow, unglamorous, and where most of the work actually goes. The rest of this exists to serve it.

What happens to what you tell us

This is the part that makes the rest of it work, so it is worth setting out exactly.

You write it or say it into the panel — anything. Something you liked, something you did not, a question, a correction, a thing that made no sense. There is no right kind of comment and nothing to choose from.

Then we tell you we have read it, and thank you, because you did not have to bother. Then we tell you what we are going to do, and one of these is always the answer:

None of those ends the conversation. A disagreement is a position with a reason attached, not a verdict, and if you come back at us we are still listening. Nothing here gets marked closed and filed away.

Then the correction actually gets made — in the words on the page, or in the data underneath, whichever is wrong — by some combination of a person and the machine, and you get told it happened.

That is the contribution, and it is not a small one. Most sites collect feedback and act on the loudest few. If a reasonable number of people do this, the thing improves faster and more sharply than we could ever manage on our own, because every correction comes from somebody who hit the problem rather than somebody imagining it.

Today the panel keeps your note on your own device and nothing is sent. The loop above is what the next piece of building is for.

Open questions — where your answer changes what we do

These are not rhetorical. Each one is a fork we have not settled, and a reply through the note button is a vote.

To put it the way we would across a kitchen table: this may be a stupid idea, or not; what do you think? Every word of that is doing something. This may be a stupid idea — we are not pretending to be sure. Or not — and we are not pre-conceding either. What do you think? — the floor is yours, and we have not scripted your answer.

And there are no stupid questions here. That is not a courtesy, it is experience. The question that sounds daft is usually the one nobody has bothered to answer, which is exactly why it is still sitting there unanswered. Yogi Berra spent a career saying things that sounded ridiculous and turned out to be precisely right — nobody goes there anymore, it’s too crowded is a complete account of what happens to a good thing when it gets popular. So when something you want to ask sounds silly to you, that is worth a second look before you decide not to send it. It is at least as likely to be the useful one.

Where this page stands

A plan that does not say which parts are real is a wish list. We would rather be caught being unfinished than caught overstating.