Saving pages from your browser
The workbench's extension for Edge and Chrome saves the page you are reading into your vault, in the project and folder you choose, with a note on why you kept it. Filing a page at the moment you save it is what keeps a collection usable: a page left to be sorted later is, in practice, a page never sorted.
The page is kept as it was open, and the workbench takes the article out of it when it reads it, without the menus, banners and sign-up boxes around it. A PDF open in the browser is saved as the PDF.
Built The extension is part of the local workbench, which is not yet released. The cloud workbench will take pages from it too, once you can sign in to it.
Adding it to Edge or Chrome
Once, on each computer. It is not in the browsers' stores yet.
- Start the workbench once. It tells Edge and Chrome where it is each time it starts.
- Open
edge://extensions(orchrome://extensions) and turn on Developer mode. -
Choose Load unpacked and pick the
browser-extensionfolder beside the workbench. Its Settings page, under Saving pages from the browser, shows where it is. - Pin the extension to the toolbar.
Saving a page
Press the button, or Alt+Shift+S. Pressing Save at once always works.
- Save to
- Every project and folder, found by typing part of its name. The extension remembers where the last page from each site went, so a second page from the same journal goes to the same folder unless you choose otherwise.
- Note
- Why you kept it: the thing that is obvious now and gone in a fortnight. It stays with the document, and you can change it there.
- Tags
- Words to file it under, separated by commas; the project's own tags are offered as you type.
The popup says what it will keep before you press Save: the page as it is open, the PDF the tab shows, or the PDF a page links to.
Papers behind a publisher's reader
A reader is not the paper.
Some publishers show a paper in a reader of their own, which draws it a few pages at a time as you scroll. Saving that page would keep only the pages drawn so far. So on an article's page the extension looks for the PDF of the same work — the address the page gives for it, or a link to a PDF with the same DOI — fetches it with your own access, as the page's Download button would, and saves that. The page itself is kept only for what it says about the paper: its title and its DOI.
Where the PDF cannot be fetched, the extension does not save the reader in its place: it asks you to open the PDF itself and save that tab.
A paper that names its DOI is filed under the reference your project already has for it — one your Zotero library brought in, say — rather than recorded a second time.
When the workbench is closed
Saving never waits for it.
A page saved while the workbench is closed waits in the browser, and the extension's button shows how many are waiting. They go in once the workbench is open, and the popup can open it for you. A page the workbench refuses stays in the popup with the reason, to try again or discard.
What it can and cannot do
Nothing leaves your computer.
The extension talks to the workbench through the browser's own channel for talking to a program on your computer, not over the network. The browser lets only this extension use it, and the workbench opens no port that a web page could reach.
The extension can ask for three things: whether the workbench is there, the projects and folders to file into, and to save a page. It cannot read anything in your vault.
What it reads, where that goes and what it keeps in the browser are set out in its privacy policy.