Tools

Deterministic tools that state what they cannot tell you: a honey finder, a comparison table, a substitution calculator and a flowering calendar.

How these work

Every tool here is deterministic: the same inputs give the same output every time, and the logic is plain code you could read. Nothing is scored, weighted or personalised, because a reference site that cannot explain why it recommended something has stopped being a reference. Each tool also states what it cannot tell you — that is a required field on the record, not an afterthought.

Available now

  • Honey finder

    Filter the catalogue by intensity, colour, crystallisation behaviour and what you want to do with it.

    Answers: Which honey should I buy for what I actually want to do with it?

    What it cannot tell you
    • It filters what is in the catalogue, which is a fraction of the hundreds of named honeys produced worldwide. A honey not appearing is not a judgement on it.
    • Sensory fields are editorial assignments, not measurements. Two people tasting the same honey will not always place it in the same intensity band.
    • Crystallisation and composition figures are typical ranges for the type, and any individual jar can sit outside them.
    • It cannot tell you whether a honey is available where you live, or what it costs.
  • Honey comparison

    Put two or three honeys side by side on the fields that actually differ.

    Answers: How do these honeys actually differ, beyond colour?

    What it cannot tell you
    • It compares stored fields. Two honeys can share every field on this table and still be obviously different in the mouth.
    • It cannot compare specific jars, brands or vintages — only honey types.
    • Fields marked as not recorded are genuinely unknown to us, not zero and not unimportant.
  • Honey substitute calculator

    Convert between honey and sugar in a recipe, with the liquid, temperature and browning adjustments that conversion actually requires.

    Answers: How much honey do I use instead of sugar, and what else has to change?

    What it cannot tell you
    • Baking is chemistry, and a conversion is a starting point rather than a guarantee. Delicate recipes — meringue, some sponges, anything relying on creamed sugar for structure — may not convert at all.
    • It cannot know whether your recipe is chemically leavened, which changes whether the bicarbonate advice applies.
    • It says nothing about nutrition. Honey and sugar are both free sugars, and substituting one for the other is a flavour and texture decision, not a health one.
    • Volume measures for honey are unreliable because honey clings to the spoon. Weigh it.
  • Flowering calendar

    See which plants in the catalogue flower in each month, and find the gaps in your own garden's year.

    Answers: What should I plant so that something is always in flower for bees?

    What it cannot tell you
    • Flowering dates are northern-hemisphere temperate defaults. Records carry a hemisphere, and southern-hemisphere readers should shift roughly six months — but that shift is an approximation, not a translation.
    • Local timing varies by weeks with latitude, altitude, aspect and season, and blossom dates have been moving earlier.
    • It shows the plants in the catalogue, not everything worth planting.
    • Flowering is not the same as yielding. Nectar secretion depends on weather and soil, so a plant can be in full bloom and offering very little.
  • Pairing finder

    Start from the food or drink rather than the honey, add a flavour or strength preference, and get the honeys that fit with the reason each one matched.

    Answers: I am eating this — which honey goes with it?

    What it cannot tell you
    • Selection is structural, never textual. A honey with a lyrical note about tea will not appear on the tea pages unless its recorded intensity and flavour families put it there — which is correct, and does mean the tool sometimes disagrees with a description you can read on the same site.
    • Pairing is a matter of taste, and this tool models none of yours. It narrows a catalogue to a defensible shortlist and then stops.
    • The rules describe honey TYPES. Any individual jar varies by season, region and producer, and so does the food it is going with.
    • A honey missing from a list may simply have no note recorded for that food rather than being unsuitable. Absence here is weak evidence.
    • It cannot tell you whether a honey is available where you live, or what it costs.
  • Seasonal bee planner

    Pick a season and a goal, and get the plants that flower then — with the months each covers and the gaps the list still leaves.

    Answers: It is this season and I want to do this — what should I plant?

    What it cannot tell you
    • Seasons are northern-hemisphere temperate. Southern-hemisphere readers should shift roughly six months, and that shift is an approximation rather than a translation — day length and rainfall do not move with it.
    • Nest sites limit wild bee populations at least as much as flowers do, and a planting list does not address them. The output says so.
    • Flowering is not yielding. Nectar secretion is far more weather-dependent than flowering is, so a plant can be in full bloom and offering very little.
    • It does not filter by region and will not guess one. Matches carrying an invasiveness note are flagged instead, because recommending ivy in Oregon or gorse in New Zealand would be actively harmful.
    • It cannot know your soil, and soil pH decides whether a good many of these plants will grow at all.
  • Flavour selector

    Start from a flavour you already like and work outwards to honeys that share it.

    Answers: I like this taste — which honeys have it?

    What it cannot tell you
    • Flavour families are editorial groupings, and the boundaries between them are matters of judgement rather than fact.
    • It does not model intensity of a family, only its presence — a honey with a trace of smoke and one dominated by it both match 'smoky'.
    • Adjacency is co-occurrence in this catalogue, not similarity of taste. It answers what tends to be in the jar together, which is a different question from what tastes alike.
    • The map shifts as the catalogue grows, which is correct and does mean a link shared today may show slightly different neighbours next year.
  • Storage guide

    What to do with a jar that has gone solid, cloudy, or has been sitting in a cupboard for three years.

    Answers: How should I store honey, and what do I do when it sets?

    What it cannot tell you
    • It cannot assess a specific jar. Anything with an off smell, visible mould or active fermentation needs a person to look at it.
    • Crystallisation figures describe a honey TYPE. Any individual jar can behave differently depending on how it was handled, stored and filtered.
    • Reversing crystallisation with heat degrades aroma and raises HMF. The tool says so rather than treating it as free.
    • It assumes the jar contains what the label says. Adulterated honey behaves unpredictably and no storage advice covers it.
  • Bee-friendly garden planner

    Build a planting plan for a specific space that covers the whole season and suits more than one kind of bee.

    Answers: What should I plant in the space I actually have?

    What it cannot tell you
    • Nest sites limit wild bee populations at least as much as flowers do, and a planting plan does not address them. The output says so.
    • Plant suitability is regional. Several plants here are seriously invasive outside northwestern Europe, and the tool flags them rather than filtering by a region it cannot know.
    • Greedy selection is not the mathematically optimal cover. It is chosen because every pick can be explained in one sentence, which an optimal solver could not manage.
    • Bee-group counts come from the documented relationships on each plant record, which are the well-attested ones rather than an exhaustive list of visitors. Treat the number as a floor.
    • It cannot know your soil, and soil pH decides whether a good many of these plants will grow at all.
  • Plant recommendation tool

    Find plants that serve a particular bee, or fill a particular month, in conditions you can actually offer.

    Answers: Which plant fills this specific gap?

    What it cannot tell you
    • Bee-plant relationships in the catalogue are the well-documented ones, not an exhaustive record of who visits what. Treat an empty bee filter result as a gap in the records rather than in nature.
    • It cannot know your soil, and soil pH decides whether a good many of these plants will grow at all.
    • It does not filter by region and will not guess one. Matches carrying an invasiveness note are flagged instead, because recommending ivy in Oregon or gorse in New Zealand would be actively harmful.
    • Flowering windows are northern-hemisphere temperate and are a guide rather than a schedule — nectar secretion is far more weather-dependent than flowering is.

Planned

Listed with what they will do and what they will not be able to tell you. A "coming soon" card with no content is a promise rather than information.