A personal link mill | Hacker News Reader

A personal link mill

corte.si

71 pointsjoeyespo17 comments

I would definitely use this, and probably (depending on just how much it improved my life) pay to use it.

One tip, to solve the "chicken and egg" problem of getting a good garden going while still consuming current levels of information, allow the user to choose to include some of those "low quality" sources as well.

For example, right now the vast majority of my info comes from HN and Reddit, both from links and comments. Ideally I'd be able to subscribe the service and keep getting all that info, then as I slowly build up my own more personalized, curated sources I could slowly start to remove the low-quality sources. I think if you could make this a pretty frictionless process you'd have a winner.

I built a really hacky version of this when I was fresh out of university and looking for jobs. Searching for jobs by keyword sucks so I just chucked a whole bunch of feeds into a database, de-duped them and presented them through a bayesian filter with a like-dislike button. It worked well enough to be an improvement on subscribing to search results in an rss reader.

https://github.com/jamii/pipes

Most of what the OP describes could probably be done pretty easily in yahoo pipes. Especially if it could call out to external services for ML recommendations.

Funny thing. I'm currently building one right now - disturbingly close to your product idea. It's currently a "generalized feed consumer" that presents a user with "stream of [de-duplicated] links to consume..."

I'm just starting to use it for most of my content consumption, so that I have a nice corpus of data about my preferences. The goal is to eventually develop a predictive model to start filtering content with...

Sounds cool, can you share a link/download ?

Sorry, not at the moment. Perhaps in a few weeks, once things have stabilized...

All the best with the stabilizing, please add me to your beta users list whenever you're ready :).

Funny, I have a personal wiki page with a couple hundred words of notes about doing exactly this. Please drop an email when you want beta users. :)

It's an idea I've kicked around for a long time too. Building a simple Bayesian system would be neat; as you read, you should be able to tag items easily, and in a way that doesn't pigeonhole you into only seeing one side of an argument (I'd like to still see the opposition's view in, say, a political discourse for instance).

I've got too much else to work on currently, but I'd love to see what you come up with :).

The google reader stack I use is to use the share button to 'save' items. Those items get added to your own RRS feed. From that feed you can do your post processing as this link did for later reading.

(edit: to see your shared items go to http://www.google.com/reader/shared/username)

I built something like this in 2005[1], got distracted[2] and still wish I had the tool today. My database was organized around links rather than feeds/items. Two stories that included links to the same thing were considered to be talking about the same thing and would be presented together as one part of the conversation. This was a low rent (but more accurate) version of what Google News does where they assume two articles are about the same subject by comparing the terms used in the articles.

I also used Bayesian classification to determine which articles were ones you were more likely to be interested in reading.

[1]: the marketing pages are still online for this ancient piece of code: http://blazingthings.com/zestynews/tour1.html

[2]: I created the TurboGears web framework while creating Zesty News, and ultimately decided to pursue the framework rather than Zesty News.

At intigi.com we're trying to solve this problem and have a beta product working. It not only helps you find interesting links in the way described in the article, but it also helps curating and sharing them.

Your site is a similar concept, but it seems to (from the little info on the front page) be much more geared toward people looking for content to share with others. I think the OP is describing a system geared toward finding content to consume for personal use.

In addition, your system seems to be more about automatically finding content, whereas the OP's system is much more manual and curated by the user.

I signed up for your beta list because I actually think your idea is interesting, but I think they solve two different problems.

This project seems alright enough for a mini hackathon. I would be interested in building such a site, however I lack some knowledge in a few things that are needed.

Subscribing to Hacker News I have figured out it being simply parsing the RSS feed to generate links. How do I go about doing it for the other types of feeds? Does Twitter and Google+ feeds require integration with their individual APIs? Any help is greatly appreciated!

Yahoo Pipes provide the fastest path to a personal link mill that I've seen.

http://pipes.yahoo.com/pipes/

Use Pipes to aggregate, filter, de-dupe, summarize, tag, and then feed your Reader.

Enjoyed the article, but how is this sufficiently different from say NewsBlur?

It would be really great if you would have some way to tie this into WordPress or any other CMS, so you can publish your personal links for the day.

good idea..corte.si... I think adding a machine learning method on reordering/prioritizing the links would be a plus...the input data for the machine learning system would be the links user clicks to read as well as links to which user clicks as favorite... or may be pulling the keywords/tags from a given link's page and then checking how many links are clicked under a certain tag will surely help ordering the links by tags...its just my idea on how to help user to read his preferred links/topics....your inputs are welcome!

Reply on news.ycombinator.com