Ask HN: For every article you read, how many do you save for later?

I've organically come to a system that works pretty well for me.

I browse HN, reddit, Zite (iPad), and sometimes click on links in my fb feed. That's my 'inbox'. Then:

1. Anything I feel like reading right away, I read. 2. an actual article, usually more long-form, I add to my instapaper list. 3. an article that might be interesting, but not right away, or anything else that is worth saving, but not useful right away, I add to Pocket. 4. for any comment pages, or articles that somehow don't go through step 2 or 3, I add them to my OneTab list (by far my most-used chrome plugin). Often these are just things I can't read right away, but I'll probably look at later in the day, or things that instapaper can't handle (HN comments). 5. anything that really stands out, I either add to my OmniFocus inbox, or to my Wunderlist inbox (Because I don't have OF on my phone).

With that done, here's how I process stuff: 1. when at my computer, I tend to first check my OneTab backlog. I open a few tabs, and read through them. This is generally during work breaks or lazy moments. 2. On the road, particularly short commutes, I read instapaper articles. I intentionally kept myself from getting internet on my phone so I'll keep from aimless browsing. Interesting snippets or articles then go to my Wunderlist inbox for later processing. 3. On longer trips, or significant downtime, I generally look at Instapaper for the 'better' material. 4. every once in a while, or when I'm reminded of something I read a while back, I check my Pocket 'dump' for the relevant material.

All in all, this kind of just started happening through use. I didn't agonize over a workflow, and I'm not sure it's optimal. But it works pretty well.

I actually have a different workload. Rather than read it, and save for later, I add my links to Pocket first, then read from there. That lets me absorb far more possibility - then read when I feel like it.

Plus, it makes the subway commute kick!

For me, there's always a 'toread' backlog. Then when I'm in the mood, I feast.

If the article is really important, I convert it to PDF using PrintFriendly.com and save it to my Basecamp. Otherwise, I save it to Pocket for a snack later.

I read somewhere by one of the developers behind those read it later apps that its not uncommon for people to have over 10,000 unsaved items never read. I'm fairly certain its pretty common behavior.

Seconded danso on Pinboard. It's simple and fast so I tend to save quite a few everyday from HN and other places.

For "articles", nearly none; I save a copy of about 80% of articles I actually read and save essentially none for reading later.

Of some high end, long PDF files, say, of books, I glance at good descriptions and save a copy if the content looks good. For the famous CLRS, I read through it quickly, saw a lot of poorly done material, but did keep a copy.

For such PDF files, my intention is to have them for reference later if needed; otherwise rarely do I have any intention of actually reading them.

For high quality books I haven't read but would like to, I have a big but old stack. But mostly those books were to help my career, and my career has taken another path. For that path, I have essentially nothing collected but yet to read.

Pinboard has been huge for me...I almost never use the HN saved feature...not because it's bad, but because it's not good at what it does...I upvote a lot of things, sometimes things I don't plan on ever revisiting but that I think are worth praise (Show HN projects)...So it's not the saving, it's the lack of filtering.

Pinboard does some sort of Bayesian analysis to automatically tag things I bookmark, but I still add tags manually. I also take a few seconds to copy and paste the first paragraph of the article. And when I need to look up articles on Ruby, algorithms, databases, etc...I just rely on my Pinboard tags

(My public tags are here...looks like I have 800+, which is not too bad for a few years)

- 4