Alex Barnett blog


How I find stuff I like

The three main methods I use to find content I'll be interested in are:

1. My RSS reader - I use FeedDemon and have done for years. I can't get through each feed every day (not even close), but I do go through them systematically. My feeds are ordered alphabetically in folders, so if I finish one day on the "G-L" folder, I'll begin at the "M-S" the next day. I've tried ordering by topics etc, but the system broke down. I "prune" feeds as I go and aim not to increase the number of feeds. If a new feed comes in, one must go that session. My main criteria for deciding which feed will go is the last updated date - if the feed hasn't been updated for weeks, it's a gonner. There's enough natural attrition using this criteria to let 1 or 2 new feeds every couple of weeks.

2. Techmeme - two or three times daily. Tells me what's hot and what's not.

3. My network - once daily. I use this page to find out which links "my network" of human bookmarkers (my people) have bookmarked that use use links as recommended reading. Some of these will point to stuff I've already read in my RSS reader (or is waiting to be read), but it works for me as a quick hit for good reading recommendations.

I have 20 people who make recommendations to me this way:

  • andyed
  • annez
  • awsbuzz
  • bobreb
  • bokardo
  • chadd
  • cshirky
  • dhinchcliffe
  • JamesGovernor
  • judell
  • Korbyp
  • ksharkey
  • leonardr
  • mattmcalister
  • ricmac
  • ryansking
  • sogrady
  • TimBray
  • twwilliams
  • vanderwal
  • These 20 people aren't making these recommendations available to me exclusively - anyone with (or without) a account can subscribe to the bookmarks of anyone else (where those bookmarks have been made "public" by the bookmarker). I didn't ask for their permission to add them to my network, but I'd say almost all of them are aware that their bookmarks are being used as a form of reading recommendations. Some explicilty use it for this purpose to republish in their RSS feeds (and / or blogs and add comments to those bookmarks (I do).

    Another interesting aspect to the network feature is that I can see who is "subscribed" to my bookmarks - who's network that I'm in. These people (some 70+ today) are using me as part of their recommendation engine.

    When I visit my network page, the "click through rate" on network-recommended content is very high. The ability for a bookmarker attach a comment to the bookmark adds another dimension too (agree / disagree, a quote or a quip).


    Michael Clarke said:

    Nice overview - I've just started using this feature myself a lot. What's also interesting is that has actually started sending traffic to my blog since I started using it more heavily - an unexpected side-effect but a pelasant one...

    # May 24, 2007 8:27 AM

    Jim said:

    You may be interested in checking out a site I recently started at This site pulls feeds from thousands of blogs and news sites, you can mark articles that do or don't interest you and it will use the content of those articles to recommend articles covering topics you like.
    # May 25, 2007 2:54 PM

    郑昀 said:

    2006年3月,我开始寻找符合中国特色的meme engine之路,很快发现只有文本挖掘算法才能做这件事情。 博客内容的文本挖掘,在中国还有一个大问题要解决。博客比新闻要复杂得多得多。2006年9月,我和中科院软件所的张俊林张博士等一起创建了玩聚网,瞄准信息过滤器和人过滤器的未来大方向。

    # June 15, 2007 4:19 AM

    旁观者 said:

    2006年3月,我开始寻找符合中国特色的meme engine之路,很快发现只有文本挖掘算法才能做这件事情。

    # June 15, 2007 4:21 AM