Hacker News new | past | comments | ask | show | jobs | submit login

I didn't think about the "indexing everything" from what they said. But now that you mention it, it does sound like a great app.

You could have an extension that indexes everything the user goes to. The trick would be making it usable; I almost never use the "history" in my browser, because it tends to never find what I'm looking for. But if you try and solve it in a centralized way, you could do several interesting things:

* Make searches across all the sites you visited work better, maybe by creating better indexes of the content.

* Make the search work according to sites you visited the most (e.g. I'd prefer the search to find comments on HN before other things, then articles linked to from HN, etc.).

Such a service might be pretty useful: browser history that doesn't suck, and is persistent and accessible online. Anyone want to build it?




I've been using the scrapbook firefox extension with an extension atop it called autosave. http://amb.vis.ne.jp/mozilla/scrapbook/addons.php

The drawback: it saves your history to local disk, and the search can get arbitrarily slow over time. I archive my history every fortnight. For older history I use grep.

It isn't something for normals, but I basically have every page I've ever browsed to since Aug '06.


"The trick would be making it usable; I almost never use the "history" in my browser, because it tends to never find what I'm looking for."

FYI: History in Safari rocks. Just open the history view and you have fast searching (including in page text - just checked) that shows all the resulting pages in a coverflow view.


Thanks for the tip, I'll have to check that out. I use Chrome and Firefox, so I wouldn't know.

Still, something computer and browser independent might make for an interesting application.


It does, but the noise is too large, and the pages will eventually be deleted, so it's not very suitable for long term storage.


I created a Firefox Addon and website for it last year that did everything you just wrote about, but I've since closed the project down. Let me know if you or anyone is interested in resurrecting it.


I'd like to hear more. Is it open source?


It's not open source. The Firefox Addon captured a visual snapshot and the entire page contents of every non-https page you visited, and stored it to an online database. The website allowed you to search your history, and results were based on number of visits to a page, length of time spent on page, number of times that page was selected as a search result, whether the page was actually bookmarked, etc. It was a very useful tool, but unfortunately the database grew large very fast. I tried to implement a subscription feature to offset server costs, but I had zero subscribers.

The code is all C#/MSSQL/JavaScript/XUL. If anyone's still interested gmail me at my username, and sorry for the off-topic posts!




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: