Top
Best
New

Posted by bookofjoe 17 hours ago

Hister: A private search engine for the pages you visit and the files you keep(github.com)
596 points | 162 commentspage 2
Yashjain413 4 hours ago|
This is pretty useful, actually. I’ve been using it for the past 2–3 days, and I didn’t find it through YC Hacker News but through some other channel. It’s great to see that so many other people are already using it and praising it—congrats on that!

One of the most useful use cases for me, especially since I work in GTM, is keeping track of new ways to get replies from prospects, whether through cold outbound or things like SEO/GEO optimization. I read at least 2-3 articles a day on this, and it genuinely helps me figure out which ideas are worth trying because I can now keep track of everything I’ve read.

jjice 14 hours ago||
Been running hister on my work machine for a month or two now. It's one of those things I don't look at until I need to, but when I do need it, it solves the dreaded hunting through the last month of browser history.

I also have it index my Obsidian notes, which is another little bonus for global search.

I did need to build up quite a few exclusion rules early on, but it's been hands off since.

sbeckeriv 15 hours ago||
I am happy to see the idea of history search more. I am on my 3rd version of my own. the use of local LLMs has made it easier to support features like weekly summarized and recipe extraction.

I like the search ui. my projects become functional but never polished. https://github.com/sbeckeriv/memoir

tombert 15 hours ago||
Interesting, I actually very recently built a similar project [1]. I was unaware of this...If I were I probably wouldn't have bothered!

[1] https://git.brucewillis.sexy/~tombert/fs_index I promise, safe for work, despite the URL.

testycool 14 hours ago|
I was fascinated by hister initially, but now I'm fixated on your URL.
tombert 13 hours ago||
Hah, I bought that URL a long time ago, mostly as an in-joke between me and some friends since Bruce Willis was my favorite actor. Then later I needed a dev URL for some work I was doing, and the only one I had free was brucewillis.sexy. Then I kept making subdomains on it for different things and eventually it just became the default.

I like it.

bradrn 15 hours ago||
Ooh, very nice! I have my own tool I’ve been using for this [https://github.com/bradrn/full-history-search/], and it’s incredibly useful, but it’s also pretty primitive. This one looks a lot nicer.
pkamb 15 hours ago||
> Your own search engine — Hister is a private search engine for the pages you visit and the files you keep.

Is there any site/project that works as a fully customizable personal front-end to all other SERPs?

When I search for something, I always want a link to the best Wikipedia result. This should always be in the same place and have a giant icon/picture.

Then there could be easily clickable links to the SERP pages for Google, DDG, etc. for that query.

A big link to route it to your favorite LLM.

Seems like you could have a really useful "homepage" for all searches that sat in front of all the other sites. It could be local only and would not require indexing the web. Also wouldn't be a files search thing, as Hister appears to be.

MomsAVoxell 14 hours ago||
I attain this without involving an untrustworthy third party, with one simple trick: Print to PDF.

Every single web page I’ve found interesting, since the advent of the Web, I have printed to PDF and stored locally for my own personal reference.

Something like 80,000+ files - my own copy of my own Internet - indexable, searchable.

Available offline. Something to read when I am far out to sea.

There is no need to involve third parties in your Internet history - no matter how trustworthy they seem to want to appear.

Print to PDF, and you’ve got everything you need, safe and sound.

computator 14 hours ago||
Not a bad idea, and I used to save lots of web pages as PDFs in the distant past, but I've noticed that a large percentage of sites will not print out correctly as a PDF these days. I see myriad problems these days: the page is truncated on the right (no matter how much you zoom out), or all pages come out blank after the first page, or the PDF gets rendered as an image (taking many megabytes per page) rather than as text, or the outer frame prints but inner frames with the actual content are missing.

The more "modern" the site, the worse it is. Surprisingly, government websites often print correctly since they've done the least amount of work to make the site modern looking.

MomsAVoxell 12 hours ago|||
> a large percentage of sites will not print out correctly as a PDF these days.

Reader mode.

jiehong 14 hours ago|||
Very true.

Even full page screenshot doesn’t always capture the non visible part of the page (below the viewport).

nottorp 14 hours ago||
Except search. I want search. Going to try this project.
MomsAVoxell 11 hours ago|||
Yeah, about search:

    $ pdfgrep -r -i -n -H "your mom" ~/PDFArchives/
Very effective, very fast, very private. Bonus points if the PDF filename itself is derived from a well formulated <title> tag, such that you can just use “ls” ..
krackers 10 hours ago|||
You could combine it with existing products like DevonThink that are meant for researchers organizing documents and provide tagging, semantic search, similarity graphs, and so on.
rochansinha 4 hours ago||
I can see using this to index the thousands of bookmarks I have and my every growing reading list and see if my local model and find a pattern and/or help surface anything interesting related to the papers I read
More comments...