I am trying out Devonthink and am at the moment very confused. Here is what I am trying to do.
I am browsing pages on my company’s network and I want to save those pages in the devonthink database to read later. If I just do a save to devonthink, it doesnt work because my companies network requires me to be logged in. This is a huge frustration.
So to get around it, I want to start browsing in the devonthink browser, login and then start saving pages. But wait! I dont knowhow to launch the devonthink browser and start browsing pages!
Can someone tell me how to achieve any of the above?
If you are using Safari and are viewing a Web page, you can use DEVONthink’s Services to capture part or all of that page to your database. (Note: if you have newly installed an application, which must be in the Applications folder, you will need to logout/login or restart to enable its Services.
To view the DEVONthink application’s Services, click on the name of the open application, e.g., Safari, in the Menu bar and click on the DEVONthink application name. Note the keyboard shortcuts available for the capture options.
For example you can select all or a portion of a displayed page and invoke the Command-) shortcut (Shift-Command-0). That will result in capturing a rich text note to the frontmost open database. The URL of the source page will be retained in the new document’s Info panel.
NOTE: Some browsers cannot access some or all Services. Firefox does not recognize OS X Service at all.
When viewing pages on a secure site, you may be blocked from saving the page, or capturing it as HTML or WebArchive.
However, you should be able to capture as plain or rich text (see above), and you should be able to “print” the page as PDF. If you are using DT Pro or DT Pro Office, invoke the Print command (Command-P), click on the PDF button in the Print panel, then choose the script to save to DEVONthinnk to…
Thanks for the help, but could you clear out a few things for me:
How do I launch the deveonthink browser? I just want to browse webpages from within devonthink.
The service menu only works for me when I highlight something. If I do Command-A and select the whole page, and then do Command-% to capture webarchive, the whole page is captured. But just doing command-% does not work.
3.The print to PDF seems a good work around, but seems to involve too many steps. Do we have a script for it?
I’ve built up a collection of hundreds of bookmarks in my Bookmarks group in my main database, organized into subgroups, e.g., scientific journals, governmental agencies, news sources and some topical collections.
I initially captured those bookmarks from another browser, as one could not directly enter a URL into the address field of the DEVONthink browser. To do that, I copied the URL to the clipboard, then press Command-N to save it in DEVONthink.
Once a bookmark has been saved, it can be renamed. For example, I renamed sciencemag.org to Science Magazine Online. When I click on that bookmark, I’m logged onto the journal, ready to check the current week’s issue for articles of possible interest.
Over time, I’ve built up a far richer and better organized collection of Web sources in my database, than in the bookmarks I have in Safari.
You can select only a portion of a page and press Command-%. Only the selected portion will be saved as a WebArchive to your database.
Yes, you will find scripts to capture as PDF, some supplied on this forum by users.
Fortunately, I’m usually multi-tasking (thinking, or parallel-processing) while invoking the sequence Command-P, click on the PDF button, click on the Save to DEVONthink to… script, so the capture process doesn’t take any time at all.
When I’m working in my database, I spend more time thinking than doing any “mechanical” work. Thinking is the really hard part of getting something done. Whether a mechanical action takes a fifth of a second or two seconds doesn’t make any difference in the outcome, so long as it’s easy and doesn’t distract me. I’m more time-sensitive to searches and See Also operations than anything else, as those are operations for which I want instant gratification; they are critical to the next stage of what I’m thinking about. That’s when I want operation time to be measured in milliseconds rather than seconds.
For me, looking at the current issue of Science Magazine, for example, isn’t much influenced by the mechanical procedures I use for capturing articles. I have no intention of capturing an entire issue. Instead, I look at the table of contents of an issue and Command-click on articles that might be interesting, to open them in new tabs in my browser. Then I work through those, one by one. Sometimes I make a quick scan to decide whether to capture an article; sometimes I read it entirely. The decision as to whether an article would be fruitful as a database capture is what takes time, and of course actually reading an article of special interest is a learning process and takes still more time. The capture process is essentially inconsequential in this workflow. I may capture 3 articles from one issue, and perhaps 15 or 20 from another issue. This mode of building my reference collections pays off. Those collections really do reflect my interests, and that has long-term payoff in their usefulness. Over the years I’ve lovingly built up a collection of tens of thousands of documents in this way.
Once in a while I’ll see the need for information that’s not currently in my database. I’ll supplement the database by searches within a journal or other source, and/or by dumping in results from a DEVONagent search – and will probably spend time in evaluating and deleting the less fruitful additions.
The workflows I’ve evolved seem quite efficient to me, using the benchmark of getting something, such as a writing project, done. Specific procedural steps may seem inefficient to others. I use a few keyboard shortcuts, and also do mousing through menus or contextual menu options. I’ve developed habits that focus on “mechanical” procedures that seem to me easy and non-distracting, things I can do on “auto-pilot” as I work through the project. I spend much more time thinking about the project itself, than in carrying out mechanical procedures.
The biggest improvement in my performance would result if I could only think better and faster. Unfortunately, there’s nothing Christian can do about that.
Does this mean that now one CAN directly input an address? The only way I could figure to do it was in the “information” inspector. Is there an easier way?
No, although I’ve seen reports by folk commenting about focus issues that it’s possible to change focus and find that the DT browser address field is suddenly editable.
I’ve built up my database collection of bookmarks in three ways: 1) copy URL from another browser to the clipboard, then save clipboard content to the database; 2) use the Bookmark bookmarklet in Safari to send a bookmark to the database; 3) When moving among links in the DT browser, to capture one, use Edit > Copy Link (Control-Command-C) to copy the link of the displayed page to the clipboard, then Command-N to save clipboard contents as a bookmark document.
Now that tabbed browsing is a feature of DEVONthink 2, I do almost all captures from the Web using my Bookmark group collection and the DT browser. The combined options of Services and contextual menu options cover the range of capture modes I may need.
You don’t launch it; Bookmark, HTML, and Web Archive documents automatically use the built-in browser in a pane or separate window.
Have you looked at the documentation, starting with the Getting Started section? And although it’s still for v1 the Introduction to DEVONthink video tutorial can help you get more familiar with the interface.
Thanks for you patience. A few more questions that puzzle me:
What is the difference between “Add page to Devonthink” and “Add web document to devonthink”. It seems that the first can be viewed offline, but not the second. If I use the first, even pages which require authentication, get saved properly. But if I use the second for the same page, I get redirected to the login page.
Why do I need to have some text selected for Command-% to work. I just want to capture the whole page.
Is there a way to map “Add page to Devonthink” to a keyboard short cut or to make it part of the services?
It seems that it is not possible to launch the browser on its own. Can we request this as a feature, as it seems almost essential.
That’s just the way that Services work – it’s a way of quickly utilizing another application’s functionality on a bit of selected text or an image or whatever. You might try clicking within the document, Cmd+A (select all), and then doing Cmd+%. It won’t be perfect, but it’s quick.
You’ve got “page” (the HTML source code) and “web document” (WebArchive) backwards. When you are offline, a page will be missing images; the WebArchive will contain the images (except for some dynamic ones).
My bank’s secure Web site will not allow either HTML or WebArchive captures. The only capture options are as selected plain or rich text, or “print to PDF”.
The WebArchive can either capture the whole page or a selected portion of the page. In either case, select.
As there is a script for this, you could turn the script into an application and name the script with a keyboard shortcut sequence.
There are contextual menu options to capture a page (HTML) in DEVONagent and in the built-in browser.
The built-in browser was never intended to be a standalone browser, but a means of rendering Web pages from HTML and WebArchive documents in a database and going out to the Web from URL links in documents and bookmarks. Over time it has improved for those purposes and the addition of tabbed browsing in DEVONthink 2 is welcomed.
Would the ability to launch the built-in browser independently of a preexisting bookmark or link be an essential feature? Maybe, maybe not. It may or may not happen. In any case, I’ve got three other browsers on my computers, each of which has features that are not likely to be included in the built-in browser of DEVONthink.
I’ve built up a large collection of bookmarks to the sites I visit in search of information to download to my databases. Most of my captures from the Web are done through those bookmarks and the built-in browser, because that’s convenient and the combination of the Services capture options together with the contextual menu capture options in the built-in browser is pretty comprehensive.
If I come across a page or link that the built-in browser cannot handle, I click in the address field and that page is then opened in my default browser. For the kinds of sites I visit, that’s very rare.
