Quick Link: historio.us
I found a great new service via Smashing Magazine today: historio.us. It’s a bookmarking service with some seriously impressive twists.
I’ve been using Delicious with Delibar for quite a while now. A while ago, I stopped archiving every web page I thought I might need someday (yes, I used to do that) because I’ve realized that Google’s current speed and accuracy have made my local data store obsolete. That’s a longer post for another time. The problem that arises when relying purely on web search is knowing what information you need, knowing that you’ve seen it before, but not being able to piece together the search to lead to that specific post or page. Delicious tags help tremendously, but nothing beats a full-text search. Limit that search to selected items from your own history, and you’ve got an amazing memory enhancer.
Historio.us lets you bookmark and tag pages, and then stores a cached version of the page and a full-text index. It’s a lot like HistoryHound, which I wrote up for TUAW a while back, but it offloads the bandwidth, processing and storage requirements to the cloud.
There are two things I’d love to see improve:
- I wish the search was a little more savvy. It does well, especially with “tags:” searches combined with keywords, but fails to handle fuzzy matches very well. Substituting “other” for “another” can lead to entirely different results.
- I wish the API would let me pull recent bookmarks the way I do with Delicious for local tagging purposes — not a huge deal.
The Historio.us bookmarklet is simple and effective, and the current API does have everything you need for creating extensions and addons for bookmarking. The bookmarklet functions much like the Instapaper bookmarklet, and you can trigger it on any page to effortlessly add the page to your index.
You can also publish a personal search so that others can search your bookmarked pages, and for paid subscribers there’s a “read later” feature that lets you mark pages as unread, much like Pinboard.
Historio.us has a free version, but it’s lacking many of the more interesting features (like Read Later). You can get the full version for $2.99/month or an annual fee of $19.95. Seems worth it to me, so I’m going to give it an extended trial and see how things go.
Very interesting post once again, Brett. But surely the difference with HistoryHound, unless I’ve missed something, is that with HistoryHound you don’t have to bookmark anything: HistoryHound just brainlessly archives and indexes everything you visit, precisely so that on that day ages later when you suddenly realise something you vaguely remember reading about somewhere on the web, on one of your browsers, may have been more interesting than you had thought at the time… and without HistoryHound you would have no way of finding it again. HistoryHound, which I discovered thanks to you, has saved things for me a number of times… historio.us sounds interesting, but unless I’ve misunderstood, doesn’t quite do the same thing.
You are correct, they serve different purposes, ultimately. Historious could extend my search abilities beyond simple tagging, which may or may not serve the intended purpose once the site is out of my memory. I don’t always come up with the best tags… having a backup full-text search is perfect.
I actually view it as an addition to HistoryHound: whereas HH indexes everywhere I surf, Historious can store a full-text record of just the things I would normally bookmark and archive foreverandever. Short-term, HH provides amazing recall, long-term, Historious indexes things you personally determine are significant (and offloads quite a bit of CPU time :).
Right: I think that puts it quite neatly. It’s an interesting addition to the world of bookmarking.
Hello! Thank you very much for your article, we’re glad you like the service! If you have any suggestions, we’d be glad to hear them. Also, we’d like to give you a free month of service (just email us to redeem it whenever you like)!
Thanks again! Stavros Team historious
Brett,
Insightful article. You’ve made me rethink me approach to archiving web content.
I’ve used Delicious for some time and love it, but as you have pointed out, sometimes it’s still hard to find that content. I am highly visual and thus Safari’s visual history is my ideal way to find recent content.
Apparently, Safari’s bookmarks capture page content as well, and since it also captures a snapshot of the page, it’s easy to find. I may start using this. I’m also considering Evernote for capturing pages.
Thanks for helping me reevaluate my workflow.
Here’s the thing. I’ve realized that a lot of us go overboard with collecting web data. We clip it to Evernote, we save it to DEVONthink, we index and tag and build huge collections. But Google has already done this! Over the last year I’ve tracked the times that I’ve needed to reference this collection of mine, and 9 out of 10 times, it’s been faster just to Google it. I do let Google store my search history, which makes my results extremely accurate when searching for things I’ve seen before. My new mantra: collect less, trust Google. The 1 out of 10 times my collection has something Google doesn’t, it’s usually out-of-date info that’s no longer relevant.
This is all part of a post I’m working on detailing how I’m actually making data collection useful for myself, and saving time to boot. It’s about using the right application for the right kind of data, and only collecting data that is going to be immediately useful. I’ll try to get that post wrapped up soon!
For the record, I’ve got my bookmarking setup scripted so that I can just bookmark via Delibar from any app, and then the tags I use determine what happens in the background (a launchd process watches the delibar.xml file for changes). If I tag with ‘evernote’, a PDF of the page gets captured to Evernote. If I tag ‘historious’, the bookmark and tags are mirrored to historious with a curl call to the API. All bookmarks are saved as local webloc files and tagged with Tags.app so that they show up in my Spotlight searches. It’s a bit complex, and quite redundant, so I’m forcing myself to make some decisions :).