docs/publishing

publishing

there is one form for everything. you give your thing a path, paste the text, and choose who can see it. that is the whole flow.

the path decides the type

bimbooo reads the extension on your path and serves the file to match. you do not pick a type from a menu; the name is the choice.

so notes is plain text, notes.md is a rendered page, and index.html is a web page.

who can see it

pictures

there is a picture link in the bar. png, jpeg, gif and webp, two megabytes each. a picture lands on your shelf like anything else, with a name and an address and the same three choices about who can see it, and you point at it from your own pages the ordinary way:

![a cat](/cat.png)

the name comes from the file you chose, tidied up, with the ending taken from what the file actually turned out to be rather than what it was called. so a png somebody had named .jpeg is stored as a png, and the two can never disagree. upload cat.png twice and the second one is cat-2.png.

pictures take up the same room your writing does. there is no separate allowance, and no album: a picture is an entry, and everything true of an entry is true of it.

two things worth knowing. you cannot change the ending when you rename one: cat.png can become hazel.png but not hazel.jpg, because the ending is how a browser is told what the file is. and a photograph carries whatever your camera wrote into it, often including where you were standing. that is taken out on the way in, and only the picture and which way up it goes are kept. if you want the file left exactly as it is, tick keep what the camera wrote in it on the form.

a drawing you write yourself as .svg is served as a picture. opened on its own, it runs no script: that belongs in an .html page, which is where it always could go. inside an <img> a drawing never runs script anyway, so nothing changes there.

putting it up later, taking it down later

under the name and the title there are two dates, and both are optional.

put it up on holds a thing back until that day. it is saved, it is yours, and until the day comes your site answers a visitor as though it were not there: not on the front page, not in the feed, not at its own address. the editor gives you a link so you can read it in the meantime, the same short-lived link a draft gets.

take it down on is the other end. on that day the thing stops being served, and it is cleared away for good not long after, so this is a deadline and not a pause.

both are UTC days rather than hours. empty either box and save, and that end of the schedule is off again.

one step back

every entry keeps the text it had before your last edit. if you paste over something good, the editor offers to put the older copy back, with the time it was written. there is only ever one copy, and putting it back replaces what is there now, so it is an undo rather than a history.

two things clear it on purpose: making something private again, and ticking "read once, then gone". if you take something down, the words you took down go too. and if your shelf is very full, a save will go through without leaving a copy behind, because saving matters more than being able to undo it.

dropping files in

drag a file from your computer onto the writing box and its text goes in, with its name in the name field if you left that empty. on a phone, or whenever dragging is a bother, the "pick one" button above the box does the same. it is only put in the box: nothing is saved until you press the button, as with anything you type.

for several at once, or a whole folder, use "a few at once" on your desk. it shows where each file will land before anything is sent, and you can change any of those names there. then they all go up together, or none of them do:

finding things again

the box at the top of your desk finds a thing by its name, its title or any word written inside it. the last word you type matches as you go, so kingf already finds the kingfisher, and accents can be left off. pictures are found by name and title only.

how often it was read

your desk shows beside each thing how many times it has been read, and its own screen draws the last 30 days, one bar a day. it is a number and nothing more: nobody's address, browser or anything else about who came is kept, and no cookie is set to count them. the day by day bars go back 90 days at most; the total is kept for good.

opening the page itself counts, and so do the raw and download links. the list of your things that your corner shows on its own, and your feed, do not count for anything. a browser that already has the page and only asks whether it changed still counts as a read.

the raw and download links

add ?raw to any address for the plain text, and ?download to save it as a file. these are query parameters, so they never clash with a path of your own.


bimb.ooo · the manual · publishing · sites · links · domains · account · what changed · about · QAF