The Blogroll I Want For AVC
Way back when I started this blog (2003), I created a blogroll. Back then it was pretty easy. There were a dozen or so blogs I read religiously. Blogs like Seth Godin, Dave Winer, VentureBlog.
Over time the blogroll grew. Then I started using a feed reader and I found one, Newsgator, that would auto-generate a blogroll. That was pretty cool. I'd add a blog to my reader and it would get added to my blogroll.
But then I stopped using a feedreader about three years ago and my blogroll got stale. And one day I removed it.
Yesterday, I received an email from Allison Kellman. She asked:
And without thinking, I wrote back:
Everything else is based on links I see on the web
I wish there was a last.fm for blogs
A last.fm for blogs! I want to scrobble my blogreading and publish it as a blog roll on AVC.
If it exists, please send me the link and I'll add it today. If it doesn't, please someone build it and I'll be your first user.
Comments (Archived):
Fred, interesting idea — Like you I found I was only reading thru links, and largely thru TechMeme and Memeoranudum (long story why) for people’s posts (vs news posts). But I realized there are a lot of posts that I don’t want to miss from people I think off as “friends” — I put it in quotes cause the list includes people I’ve never met like Nate Silver and Paul Krugman, and people I have — like you.What I came up with is a Twitter feed: http://twitter.com/friendso… — and it really worked. It’s low flow, 20 or 30 links a day — all from people I care about. It’s one of the keepers — an idea that worked. (I found this post from the feed.)
That’s a major source of the links I click on every day DaveI follow it on twitter and often go right to the page like I do with timo’reilly’s feed (which is very link heavy)Thanks for building it
Interesting. Does it make you want one of your own — FriendsOfFred?
MaybeBut then I am back to creating lists instead of following linksI’m lucky to have friends like you who do a fantastic job of curating theweb for me
oh, yea — I did something similar to friends of dave about a year ago: http://nonline.info๐ nmw
Looks like a few people have wished for the exact same thing :)http://www.google.ca/search…Dunno of anything that exists yet though!
If all blogs were using MyBlogLog, it would be fairly trivial to create a list of blogs you have visited recently or you visit most frequently.MyBlogLog effectively provides this functionality anyway, if you let it.If you allow MyBlogLog to add you automatically to communities, then the list of communities effectively becomes a list of blogs you visit.I believe they used to display the communities based upon how frequently you visited – now they are displayed in the oder you added them which is not as useful.
An alternative, is to use some kind of activity widget, and then actively use some form of social media/news voting to share stuff from blogs.It could be something as simple as shared items with Google, or something more complex such as a lifestream widget from Friendfeed or Blogcatalog or one of many others.
If it doesn’t exist, perhaps you should fund it. ;-)Pet project http://www.blogrolling.com/ has been having some issues, and I’m sure they’re smart enough to figure it out. Hmmm.Anyway, I think it is an interesting idea.~G~
Really surprised that you don’t use a feed reader Fred. This made me think about how mainstream users will find their content in the future. According to Forrester’s research, RSS and social bookmarking has very low adoption among internet users relative to other forms of social media use (they call this type of use ‘collecting.’)http://www.forrester.com/Gr…It makes me think that perhaps RSS will never actually hit the mainstream except as a protocol used by developers for syndication and that it will be the social web that notifies us of content worth reading.Or perhaps tools like http://fav.or.it could help to bridge the gap. I’m not sure if it’s quite what you’re looking for but it’s worth a look.
As efficient as RSS is, you miss the “personality” of the the blog … I like seeing what Fred (and others I read) has added. Widgets, redesign, look-and-feel … It matters. I’ve been blgging since 2001, and I rarely use a reader.Standard RSS readers are boring and not entertaining enough to mainstream, in my opinion.For Twitter, however, it is different … As it is all about text and links (at least today) you can use “readers” much more easily .. to say nothing of the fact that TweetDeck and other Adobe AIR apps are very slick and cool to use.Anyway, sorry for going off-topic, Fred. ;-)~G~
That is very much on topicFeed reading isn’t mainstreamFollowing links is
Would you say Digg does something like this? It is not a feed reader, however it is a place where people discover ‘links’ to popular stories.Is there something you would want to add to Digg?
RSS suffers from the inability to stumble upon thingsSerendipity is one of the greatest things about the webI visit new blogs I’ve never read every day and I love thatI don’t want to maintain lists, I want to follow links
Agreed. It had never occurred to me that I would do one or the other. I might try closing down my RSS reader for a week and see if I miss it.
Fred, thats exactly the pattern I have. Its awesome but sometimes it can just feel a lot like surfing around and not being confident that the time is best used. A tool that retains the “stumbling upon” feel while keeping it contained somewhat – by social graph or algorithmic recommendation – is something we are hacking an early version of. I will share when we have something that is share-able.
Please do
You’ve just defined the difference between SEARCH (lists) and DISCOVERY (stumbling). The power of your suggestion lies in finding the known unknowns (good content that you know is but a link away) rather than the known knowns in your blogroll (stuff you must curate yourself).
I quit using rss readers after twitter. Actually, i was never an active user of readers. All the blogs/bloggers i like to read everyday i follow on twitter and jump into new posts as soon as they pop on twitterfox. After all, i believe twitter will kill rss readers. I even make a post on my blog about it, too bad it`s in portuguese ๐
Diego,Then you should really checkout http://www.tweetlnks.comI was coming across the same issue. It’s all about discovery and that’s why I follow the people I follow on Twitter.
You can do this in Google Reader, go to settings, then Folders/Tags.You can make any of these items public, and then give the addresses to friends, etc.It also has a little “blogroll” or “clip” maker for your site just like you asked, and will update dynamically based on what you add and take away to the folder it represents.For example, this is my “internet/tech” folder:http://www.google.com/reade…
Yeah, but then I have to use a reader and I don’t want to do thatI want to follow links, not build lists
In Google Reader I maintain a list of my main “filters” (like you Fred)… and start my reading from there… and always go to the original blog posts when interested. Google Reader has “Reading Trends” that show my most read, shared, emailed stuff for the last 30 days.Google has to just offer a feed for my Reading Trends that I could add to my blog… and I would not have to maintain my old style blogroll.Let’s not forget that traditional blogrolls have these important social “credit” and “link juice” functions.BTW, this discussion is not new on Fred’s blog… see this (and my comment there + Joshua Schachter’s response): http://www.avc.com/a_vc/200…
Have you used the “Next” bookmarklet from Google Reader?It allows you to use Google Reader through just one link – clicking on it takes you to the next unread item, marking it as read in the process.This allows you to read each blog physically as you like, but maintain a list at the same time.
I’ll try thatThanks
Something like this should be built into Google Reader, so that they track the links you click and make that data available via an API. I see no other way this could be done. FF extension cannot be used since there is no way of knowing which links you’d want the world to see.Still, a very interesting idea and one that any RSS news reader should definitely implement.
The Trends function in Google Reader tracks your reading habits and publication statistics of your subscribed feeds. Haven’t seen any mention of there being an API for using this data externally though.Fred – Google Reader’s had quite a lot of development so it might be worth checking out again. If you want to follow links then get the Better GReader Firefox extension, which loads links that you click on right there in Reader – navigate the web from a single page! The keyboard shortcuts make it super-easy to navigate, tag and share articles, etc. Throw in the Ubiquity extension and you’ll be able to post links to Twitter on the fly as well!
It’s still reading feeds thoughI’ve moved beyond that and now just follow links
Fred -Blogrolls have often as much or even more value than the blog content itself for readers. What I really want is to get an aggregated blogroll based on the readings of my network.For example, I’m reading your blog, Howard Lindzon’s, Roger Erhenberg’s, etc. blogs. To find other good blogs, you look through the individual blogrolls, it they exist. But what I really want is an aggregated one, with weightings based on the frequency within my network. If you all read one blog, I want to know and have a look at it. Chances are I will want to read it too.Individual blogrolls can be manually or dynamically created as you suggest. But then we need an application to collect the blogrolls of my network, and present me with MY master one, ranked by frequency.Articulating a business model should not be simple, but the value created by such a service would be high and tangible. They can provide all the infrastructure (create individual blogrolls automatically, display them as widgets, enable aggregated ones) or just a piece of it (for example, just crawl the blogs I read for blogrolls and work with this).The value of an always up-to-date reading list of feeds based on the content creators I read, now and trust, would be really high.And you can think of an infinity of bells and whistles to add such as tagging the links and filtering, graphical display, etc.Up to date blogrolls are one of the rare valuable knowledge pools that has not yet been tapped. Anyone interested?
Set a Tumblr link blog. Use the “Share on Tumbler” bookmarklet and post what you like to it. Use their JS snippet to republish that there.
Too much workI love that last.fm just sits there and records what I listen toI want a client on my laptop that does the same for the blogs I visitI’ve probably got dozens of tracking cookies on this computer that do thatalreadySomeone just needs to create a database of blogs and map my clickstreamagainst that
It begs the question “what is a blog” quantitatively…. “you know it when you see it” but, is the MSM “blogs” on their news sites the same as this blog, I’d say not. What do you record and what don’t you record in your clickstream and who gets to crown you w/ the “blog” status? Do you have to be using blog software – or what if I just go old school and put up static HTML pages ๐ Hard to separate “sites you visit” vs “blogs you visit”
if a web page you visit has an RSS feed on it, count it as a “blog you visit”
There are a bunch of companies that have collected/built large indexes ofblogsI’d start there and let users add moreThere’s also the suggestion others have made about looking for an RSS feed
RSS seems to be much more pervasive than Blogs these days. As you say, I guess you could query technorati or some such:http://technorati.com/blogs…Another thought is you could look in the <head> of the page for the generator or just in general for a keyword match like “typepad” , “wordpress” , etc… that would probably get 90% then you’d have to manually add.It seems #5 on your blogrollr tweaks is exactly this issue.. I’ll be interested to see how it evolves ๐ On a side note it has been a treat to follow all this… and I found it all because you’re a “friend of Dave’s” – which make it even more so!
Fred:Such a tool would be an excellent way to measure “influence.” I love the idea and I’ll be the 2nd user.Note: If someone is looking to develop such a tool/service s/he may want to speak with Dave Winer. Dave’s OPML Manager was an initial attempt to put together such a tool.If would think that the Google Reader team would be well positioned for such a project.
Dave left a comment on this threadThe first one I think
Yes, I think our Comments crossed each other in the blogosphere.
Since I’m typing this into a Disqus comment box, that seems like the obvious starting point to me: don’t create yet another service that bloggers have to integrate and users have to register for, just start bugging Daniel about building it into Disqus.You get the added benefit that it wouldn’t just be visits–your blogroll could factor in which blogs you actually engage with (comment on) and how often you do so, versus those that you just visit.
Once again, whitneymcn, you’ve got a great idea…it seems to me a combination of disqus, twitter, and a few other services rolled up into it’s own RSS service and then widget’ized would do quite nicely for this (perhaps something as simple as parsing and making a widget out of friendfeed items that have links?).I know a number of people (myself included) are working on culling intelligent information out of links passing through twitter…and some have even started to do things that are specific to a given user ( @lmai put together an especially nice tool for this at http://tweetlnks.com ) … so there might be something there as a good starting point as well.
I like Whitney’s idea of basing it on comments. It wouldn’t be as comprehensive as the “recently scrobbled” idea, but using comments as an indicator of interest has the benefit of filtering out links you may have clicked on, but not enjoyed. I hacked together a quick proof of concept using BackType, which includes your Disqus comments as well as comments on blogs that don’t use another commenting system…http://pipes.yahoo.com/pipe…Enter your BackType user name and up to three blog URLs you want to exclude from the list (ie. your own blogs). The Pipe generates a list of the blogs you commented on most frequently based on your last 1,000 comments tracked through BackType. I’m ranking by level of engagement. If you’d prefer that it be ranked by most recent engagement, just remove the “Unique” and “Sort” modules under “Edit Source” (Fred, I think you have some experience with Pipes).You can display the output as a badge using the standard Pipes widget or you could grab the JSON output and skin it to match the design of your blog. Of course, this is just a quick hack to demonstrate the concept.
Is there anything that Joe Lazarus can’t do with pipes? :)I’d love to see it extended to visits, though, which could be done as a part of Disqus (or similar). It seems to be a split these days between people who are still feedreader focused and those who aren’t, and oddly (i.e. not at all what I expected a year ago) I fall into the latter camp.I still use FeedDemon (best feedreader ever), but I hit as many blogs through links coming from Twitter or other sources as I do from RSS. Where a year or so ago it was relatively uncommon for me to read a blog post on the Web, I now end up hitting actual blog posts fairly regularly, so I’ve probably got a meaningful amount of “presence” data these days.My ideal would be a combination of feedreader data, site presence, and comment activity, but if I can’t get all three of those the latter two feel most significant to me now. Hmmm…anyone else interested in seeing a Newsgator/Disqus meeting of the minds?
Thanks, Joe. This is very cool. It’s also something we could make available very easily via our API (by username or URL). On user profiles, we show a “blogroll” based on where a user comments (limited to 50 or so blogs). For example:http://www.backtype.com/fre…We’ll make a widget right now that shows up to X blogs sorted by activity. Fred, is that something you’d be interested in?Update: The widget is available @ http://www.backtype.com/wid…
Awesome I’ll check it out
That seems to make sense.I guess this would need to work as a browser plugin, you’d click to scrobble. delicious wasn’t a million miles from here either. They’re both missing the post recomendation functionality though, you’d need to mash stumblupon, techmeme, alltop and what friends are reading into the mix for that.
Yes, that’s what I have in mind
This is what Toluu does in its most basic use case.
I’m surprised nobody has mentioned Toluu: Toluu was designed for exactly this. Sadly, it’s not getting the attention it deserves. http://www.toluu.com/ . The design is a little awkward but the “seed” is there. A last.fm for blogs. With a special bookmarklet provided by Toluu, everything you subscribe to gets added to your Toluu before it’s added to your favorite feed reader.
That’s a step in the right direction but I want something that doesn’trequire using a reader
Fred, you don’t even have to use a reader. For reach feed (say http://www.toluu.com/feeds/… ) you can just click “Add Feed”, and it will be added to your Toluu list. The feed reader integration is optional. If you select a feed reader, the “add feed” page will redirect you to your feed reader, but if you don’t select a feed reader, Toluu will only act as a database.Maybe you should talk to Caleb: http://friendfeed.com/caleb… (Caleb is the founder and lead developer of Toluu)
Toluu can also generate a Tweet when you add a feed. Like this you can share your favorite blogs with your twitter following.
You know what’s funny? Because of you I see a whole new use-case for Toluu. I always wanted it to be an exact replica of what I had in Google Reader (and it fails in that, because it lacks syncing options). But just as a way for “scrobbling” feeds, it works perfectly. If you don’t want the Twitter integration, you can add your “toluu feeds” rss feed to Friendfeed. It works much like last.fm “loved tracks” then I think. With the exception that you can’t add a feed to your Toluu account twice I think… You’d have to remove it from your Toluu list first.
I don`t think toluu.com is exactly what Fred wants.I believe i understand what you`re thinking Fred, such thing could be easily built as a browser extension. Actually i’ve never developed a browser extension, but i have some friends that did. I`ll ask them for some help and then i might be able to come up with something soon.
ThanksAs andrew said in another comment, you could leverage the attentiontrust recorder extension to do this
Heh, that was my thought too. I wrote the attentiontrust recorder so long ago, I’m not sure how compatible it will be with Firefox 3, but it might not be too hard.The tough points would be:(1) What counts as a blog post? — I think you can do pretty well with some combination of (a) by looking for a RSS feed (b) common blog post URL patterns (c) common blogging domains (blogspot.com, wordpress.com, etc.) (2) Where to store the information? — The fasted hack would be to send each link to a twitter feed.Seems like a fun little project. If I have time this weekend I’ll give it a shot.
Woot
Btw – it would be great if I could manually add domains that aren’t recognized as blogs
Fred, I don’t get your RSS objection. Dedicated readers are a pain, but if the “linkroll” were stored in RSS doesn’t mean you (or your visitors) would have to read it in an aggregator.I think of RSS as a normalized message format. There are plenty of widgets and add-ins that convert feeds into something for web sites, for email (e.g. FeedBlitz at over 2.5 million messages / day), IM, Tweets etc. Is it RSS you really obejct to or just aggregators as RSS client?
I love rss. It has facilitated a revolution in media.What I don’t like is reading feeds in a dedicated readerI like to read in a browser on the web with all the personality that each blog brings
Give Google Reader another shot )))Gotta love the “J” key!
I’ve triedWhen is enough enough?
I use Google Reader to give me the topic headings as soon as they’re published.If I like a heading, I click on the link and always read the post in its native blog page.The only problem is that every time I stumble upon a blog I like, I add it to the reader which means that sometimes when I sit down at my PC I’m faced with over 400 new entries, which is a bit daunting.So, all this boils down to organization and authority again. What we need is an intelligent topic/authority based filter for blogs, just like we need it for Twitter.
This is specifically the reason i hate all rss readers except Netvibes! In netvibes there’s a button that says “show website”. It opens the actual site to the page you wanted to read within the window but keeps the feed to the left. Try it out but make sure to press the ‘show website’ button.
What are the folks at Xobni up to? Sounds like ‘Xobni for browsers’ is what you want. Sort by activity / time, use some combination of hard and heuristic thrsholds to filter out drive-bys and I think that’s it.
The closest thing to that is reading blogs on google reader and marking everything you read as share. But that’s not quite what you wanted…
Delicious, Tumblr and Google Reader all have bookmarklets for adding to their systems. Google Reader has a “share with note” bookmarklet that lets you add any link to your shared items list (even if its not in a feed in Reader).However these all store individual links, without providing stats on aggregate sites. Reader’s “Trends” section shows which subscriptions you have shared the most, however I’m not sure if it includes sites that aren’t in your feed reader. Aggregating site information from links could be programmed by querying the Delicious API, or rss feed, and calculating the most popular sites.A Firefox extension could do this by monitory the urls you visit and bubbling popular sites o the top. Google Chrome does this with the “Most Visited” list, but that list isn’t exposed in an accessable way.
The latter approach is directionally what I’m talking about
Here’s a proof of concept written entirely in JavaScript hitting the Tumblr JSON API:http://monsur.com/tumblrsit…It grabs a user’s last 50 links and aggregates the top sites. The code is rough around the edges (it was about 10 mins work, tested only in Firefox :-). But it could be polished up and dropped on pages as a widget.
Cool. I’ll check it outThanksfred
Fred, isn’t there a fundamental problem with any monitoring feature that tracks ALL URLs that you are reading as opposed to the ones you find useful? The reason twitter links work is because (except the automated posts) they are somewhat hand picked and recommended by people you follow. So the exact last.fm model may not be that useful / possible.Also, I see that a lot of the comments are focusing on the “blogroll” word whereas I understood your real need as something that allows you to discover good links to read and also not miss the ones that you do like to read regularly?
Well the origin of the idea is to dynamically produce a blogroll
Scrobbling blog reading reminds me of the old AttentionTrust recorder. Someone could be the blog scrobbling app for that recorder a save a bunch of time in implementation.But, at the end of the day, I think that would be a very geek/niche use case.
The share option in Google Reader does just that, its a bit clunky though:http://googlereader.blogspo…
So when do you think Google will allow us access to this data via a Google Reader API? (assuming for a person using Reader).
Fred, You already sort of have it. If you ever commented on a blog entry that you have it at http://www.backtype.com/fre…With 5000+ comments you do have a blogroll, now if you can get backtype to make it available as a widget than u have it!It wont be comprehensive but it will have quality!
We just created a widget for this:http://www.backtype.com/wid…It looks like all our other widgets, but is customizable via CSS.
There are some blogs I read very frequently but never comment onSome tumblogs I love don’t even have commentsSo while I agree that this is really useful info, it’s not perfect
I agree! A very simple way to do it would be through a browser plugin, but privacy concerns exist (atleast for me).I wonder if any analysis of backtype + twitter links (posted + favorites) + delicous + friendfeed likes would do the trick!The underlying theme that is emerging is this, we have all given our data in the web20 era & now we are demanding information back. I wrote about it at http://arjunram.com/2008/12…Thoughts welcome!
Yup. I want my data (not back but at least shared with me)
There’s an interesting tension here: the easier it is for you to scrobble or share links to things you reading, the more noise you will create. The more work required (eg, going to the trouble to email or tweet a link), the better a filter you will be. I suppose the opportunity here is to make it very easy to scrobble and then use some algorithmic filtering to reduce the noise. For example, click a bookmarklet to scrobble something, and then in the view only display things that 2 or more of your friends have scrobbled.For something easy, you could just use delicious or bitly bookmarklet to one click scrobble things: your blogroll would then be your bookmarks or bitly history ( http://fredwilson.bit.ly/ ). Maybe the more intelligent filter will come soon.
KortinaI don’t want to list every blog I visit. I just want to compile top ten inthe past week or month (or both)That creates the signal out of the noise in the scrobbleLast.fm in theory has the same signal/noise issue, but when looking at thetop lists, I get great results
Hmm, I see. So you’re looking for the equivalent of “top bands” as opposedto “top songs.” That is a pretty simple problem I’m sure one of the talentedreaders here could whip up a solution to in a day (if nothing like thisexists already).As a consumer of this type of data, seeing your list of top blogs doesn’thelp me so much because I really don’t want to subscribe to another rssfeed. What I want as someone seeking curated content from people whoseopinions I respect is a list of particular articles to read. Eg, supposeyou, john b, and josh k all read a particular article. If you are all insome network I’m following, that article should bubble up as somethingimportant for me to look at.I have a hunch the top content vs. top publisher model would result inbetter signal/noise ratio. It is however, a bit harder problem, and itrelies on some social filtering, whereas your blogroll idea only requires asingle person (you) inputting data to create a list of blogs I should read.
I’ve got a blogroll that does what I want up on avc now that was built by one of the readersIts not perfect yet but its a startI blogged about it this morning
The rapidity with which this came into being is exactly why I love theinternet. Very cool.
I would say this is a great project for the team over at Zemanta to start exploring.
Digg has a recommendation engine. Based on articles that you vote for it will recommend other articles that you may be interested in. Is this what you are looking for?
Not trying to be all ‘you vs us’ here ๐ but reddit has a sub-reddit feature that would be perfect for “a last.fm like blog recommendation.
No, I want to list the blogs I read most frequently, but I want thatcalculated for me and I want it to be dynamic and change over time
http://twitter.com
Coupla thoughts.I think this was mentioned already. I installed the delicious FF plugin and when I like something, I bookmark it. The RSS feed from delicious is hooked into my tumblr account, so it gets published there.There’s also Social Browse. I completely forgot about this Y Combinator company. It sounds like they’ll give most of what you want sans an easy-to-install WP plugin. But they do have an RSS feed that you could hook into tumblr or another WP plugin that parses RSS feeds.HTH,Rick
Here’s another option, too. Read It Later is FF plugin that I use to mark things I want to read later. The cool thing is that I can also download all the things I want to read for offline viewing — when I’ll be away from wifi for a while. It also has a RSS feed that you could hook up to Tumblr or WP. Though, again, unfortunately it doesn’t have a ready-made WP plugin. :-((And, I’m thinking you use WP as your A VC platform. If I’m wrong about that, apologies.)
I use typepad but I knew what you meant
Web scrobbler? Sounds to me like RescueTime.
Hi Fred, there is a web 2.0 company in China doing this in beta for a while, and has just launched the product officially. The company started its business with book reviews and recommendations (Amazon China didn’t launch the “books you might also be interested” function till recently), and then it adds SNS and do other experiments, such as the blog recommendation system. The blog product is called Nine O’clock, which allows you import your RSS and then it will generate blogs and blog articles that you may be interested in.It’s all in Chinese right now, and has some problem in reaching blogs out of China due to the government Great Fire Wall. You may take a look at via http://9.douban.com/. The main site is http://www.douban.com.
To automate blog submission to Toluu, one could write a Greasemonkey script that triggers the Toluu bookmarklet code (really simple) in the background. Again, only thing that wouldn’t work is “scrobbling” the same blog twice. But I think Caleb could easily change that in his schema.
Fred are you using windows or mac? I have been thinking about blogs for sometime and recommendations. I can build a simple client to try it out.
I use bothWindows in the officeMac everywhere else
I think Outbrian with their newly raised $12M has the potential to build something just like that! But then again it’s not an immediate solution you’re looking for!
Google Reader offers a “Next โบโบ” button (a JavaScrpit hack available through Settings – Goodies) that let you jump from the original blog page to the next post, without using their bland interface. That way, you can have all the design galore, the comments and Flash embeds โ like the (blog) Creator intended.My guess would be to subscribe to your Twitter RSS feed, Techmeme and Delicious Best-of, a few pundits and Scoble’s Shared, and you should be covered.There is a relevance algorithm available, but to make it yours, you need to provide feedback: I’m assuming your browsing pattern (send by Google toolbar) or the “Sharing” feature might help โ I’m not sure there is any off-site JavsScript hack available for that, though.PS: Had to type this comment three times: Firefox/Safari crashed on your page. I tried many combos, reboot, etc. โ but with nothing but your page on, whatever the browser, it stalls. With the Widget gallore, it might be hard to pinpoint the problem, but if other’s had the same issue, it might be worth investigating.
Thanks for letting me knowUsually when I have these issues, itโs because of some funky flash videoplayerIโll keep a lookout for more of these issues
Wow what a brilliant idea! I’ve always wondered something similar but about RSS feeds in general… which is why are they publisher specific when I really want vertical specific. One feed of the best web 2.0, vc, soccer, music, etc. articles at any given time. I understand the inherent subjectivity of reputable “sources” etc. I just wondered if it could ever be done.BTW – you’re site is one of the few on my blogroll.
GREAT idea. Even if I’m pretty static in my readings.
Kind of like the RSS feed from Google Reader, but independent? I wonder if there are some Greasemonkey scripts.
Hey Fred– Tony from RescueTime here (thanks Lionel for posting and thanks to BackType for lettin’ me know!).We could whip that up in about 30 minutes using the technology we already have. It could basically be a 3-tab widget that says, “What I’m reading this week” (with tabs for today and this month). The blogs/news sites would be root urls rather than full blog entries and they’d be sorted by the amount of time you spent at that domain. It’d show only sites that our service (or you!) have categorized as news/blogs. Heck, we could show the amount of time you spent at each site, too.The side effect of this is basically that you’d have to install a light app (Mac or PC) and be storing pretty much second-by-second attention data at our service. We have a mess of privacy features that can limit what data we collect if you so desire and you can delete any (or all) data.We currently have over 100,000 users and we’re a funded company (YCombinator and True Ventures).Anyhoo, if you’d like us to whip it up for you to try out let me know. Heck, if anyone else reading this would love to have something like this, chime in. If you prefer email, tony at rescuetime dot com.If anyone else wants to build this (as a browser plugin) using the RescueTime APIs, also feel free to let me know.
That’s what I want. Others seem to want to build it too. So if you do build it, pls let me know
I recently had a similar (although using a different sourcing model) idea about transient blog reading.Would you want it to automatically track what you clicked on (passively?) or would you rather it be an active choice of to include it into in your reading list / blogroll? My initial implementation idea was a bookmarklet, so you’d need to actively share individual items.When I had this thought before (with a more passive tracking model), the one thing that popped into my head is, how do you distinguish between something you clicked for reading vs. something that’s a utility your using (many of which have RSS feeds). There’s also the problem of clicking on links and reflecting off if you didn’t find the source interesting enough. Of course I tend to over engineer things so maybe people don’t care. :)If there’s really interest in this, I could whip something up on our (Grazr) feed infrastructure relatively quickly, maybe this weekend if I have some spare time.
I’m with you mike, a passive tracking model will make things much harder. I would have to turn it off and/or remove ones I don’t want to scrobble after the fact.I’m in the camp of clicking a button or keyboard shortcut to add to your scrobble.
I think you start with an index of blog domains and then let users add to their own index
I like the recommendation from @whitneymcn to use BackType or Disqus to form the basis of that index. Starting a whole new tool from a firefox plugin could create a lot of extra noise, as it’ll be tough to define which ones are blogs.Maybe something WebMynd – http://www.webmynd.com/ could just look at your stream, and pick out the blogs that you are visiting according to the index of blogs you get from BackType or Disqus?
Also curious what made you eschew a feed reader. I think this is going to be the rule rather than the exception over time.
I like to read on the web with all the color and character of the blog in full viewAs to what will be mainstream, who knows?But I do know that rss readers are less than 50pcnt of this blog’s readers (usually around his 30s)And I’ve read that less than 10pcnt of all blog readers use feed readers
Fred, I was wondering what you’re thoughts on alltop were?
I don’t have any to be honest. It hasn’t worked its way into my daily habits but I does seem useful
In other words, you’d essentially want an application that publishes your browsing history, with the only additional kicker (if I’m reading your last.fm analogy correctly) being that in order for a site to get published, you’d have to actually remain on the site for a set period of time so that it could be properly “scrobbled”.Seems like this would be a great firefox plugin, and would immediately replace the value of a blogroll as it would reflect what guys like you (trusted sources of information) are currently following, as opposed to simply listing the sites you once found worth your time.
Bingo. But I want it limited to blogs and other informational content (like msm news)
A firefox plugin could do this pretty easy, and to limit it to blogs, you can start with the basic criteria that the page needs to have an RSS auto-discovery tag on it…Another alternative might be hidden somewhere in Google Web History ( http://www.google.com/history ). Not sure if there’s a data API for it, but I’ve I’d love a better way to search/browse through what I’ve read/opened throughout the day. I’ll often click links open into new tabs, then get sidetracked and never get to them, so something like this would be helpful from both the social and selfish perspectives!
Good ideas
sorry if i’m late to this post, but how about del.icio.us?
They could do this pretty easily
Unfortunately, this site doesn’t filter out blogs from the rest of the sights you go to, but check out TimeLope, It seems to be the closest thing to what you want.
Yeah this seems pretty similar
I think you should suggest to Alex Iskold that he automatically generate this for you. It should be easy for him to add into Glue. I suggested something similar recently for delicious / Glue integration.
I was thinking the same thing. I’d like to see Glue get personalized that way. I get the top down but top down can still be based on my network.Speaking of which i wrote a post just yesterday called “My Network Is My Search Engine”. Now how to harness that in a web service? That would be awesome.
Wow – looking at the comments here, FriendFeed and I am sure there are some on twitter, I wonder if you need a global system for monitor, replying and managing comments ๐ Backtype may be a start but not quiet the right solution.
This is a big problem for me. I get comments here, friendfeed, seeking alpha, twitter, and sometimes facebookDisqus has done the friendfeed connect and I hope they’ll do tweetbacksAnd alley insider swaps out their comment system for my comment system when they rerun my postsI am hoping seeking alpha will do the sameNot sure what to do about facebook
Thanks for confirming on this – I was just wondering if there was a solution/hack that I was not aware of. Its a tough problem to solve but also key in building the social media mindshare – typically its the “conversation” and not just a single comment that delivers value (like you have here on this post) and if there’s no global way to track “replies” then chances are we will miss building that conversation in most cases.
Totally agree
Theres definitely a demand for people being able to @ people or # conversations across all these services so you could have one global conversation.
That’s a good point you make Fred that I hadn’t realized. I still skim Google Reader a few times a day, but it is mostly a practice of read it already, don’t care, don’t care, read it. Anything I would care to read I would have seen somewhere else throughout the day, be it twitter, or a link from somewhere else etc, the actual reading of my feeds is largely unnecessary now. Maybe I’ll try and phase it out.
Hi Fred,If you tweet the URLs you read, you can see them here: http://twitturly.com/user/f…The next version of our site will have RSS for that page, and we’ll have a widget you can put on your blog shortly after that.
It sounds like you’re thinking of a passive Delicious?On a tangent: Have you looked at Instapaper/Give Me Something to Read?Instapaper is a bookmarklet from one of the tumblr developers to save pages to read later.Give Me Something to Read takes the most saved pages every day and publishes the three or so most interesting sounding pages.
Yes I’m familiar with those tools. Great stuff
Great idea! Like you, I no longer really look at my feed reader. It was overwhelming and I felt guilty that I couldn’t ready them all so I pretty much quit using a feed reader. The posts from blogs I read regularly come to me in my email. It’s amazing to me that there are blogs out there where you can’t easily subscribe in email! There are a few I would like to follow, but if I have to go somewhere other than my inbox, it won’t get read. If it’s in my email, I scan the headline and first few sentences…if my interest is piqued I’ll read the whole thing…if not, I hit delete.I also follow links that people send me directly and in twitter if I think it would be interesting.Seth Godin’s blog is one of my favorites too! I like his short and easy to read/digest posts. Often it seems like he is speaking to me directly.
Off-topic, but I find it ironic that the Feedburner ad at the bottom of your feed reads:Environmental Capital Partners. The way it’s placed it actually looks like your signature banner at first glance. “What, Fred Wilson has changed firms?’:-)
Creative advertising! I wonder if it works
How does FriendFeed fit into this? I had an idea similar to FriendFeed about a year ago, but then couldn’t figure out how to get it off the ground. Oh well.
Friendfeed is like a feed reader on steroids in my mindWhat I am looking for is a way to reflect that I don’t use a reader
so you basically want to publish your web history? they way i envision that is a toolbar button you could whack whenever you want to add a page to your so-called blogroll
I don’t want to push any buttons. I just want to record. And I want the index limited to blogs and other news/information sites
You’re a brave man Fred. I’d be pretty weary of having an auto-scraper publish every blog and news site i visit. BTW, great site, really insightful. Look forward to reading every day.
At one point I was publishing my entire web clickstream via the attentiontrust recorderWe live in public so we might as well accept it
I understand this. I use a site called http://www.socialmedian.com who i think would be the perfect candidate to do something like this (they are like delicious but specifically for news and blogs).I think if you are going to autoshare, you need a list of relevant sources. You could use an index of blogs, but that still leaves out the option to manually add something of interest. Thats why I recommend a news discussion/blog sharing site like socialmedian.I posted a comment on their getsatisfaction page. If anybody is with me, click ‘I have this problem too’ and maybe they will do something about it. http://bit.ly/eaeD4
Amazing to see this post and ensuing comments evolve. I was imagining a conference room full of people sketching out ideas on a white board. Would not be surprised if 5 solid apps come of this, 1 of which is actually funded by Union Square. Really cool.
Well see about the funded by usv part. But I agree about the first part. I’ve already recvd a few apps to try out
A few thoughts:1. Ambient News is a Firefox plugin that a/ watches what you read (browse) on the web, and b/ develops a page of recommendations for you as new posts/pieces emerge on those sites (it’s watching the RSS without you making a list). Ambient News opens in any new empty Firefox window. It’s conceivable that Atul Varma, the guy behind Ambient News, could be convinced to develop a ‘streamroll’ for Ambient News.2. I use Snackr as a replacement for traditional RSS readers. It’s very different for a number of reasons: see http://bit.ly/Q7wtz. Most important, it brings back that serendipidy, since I can have hundreds and hundreds of feeds, but ignore 99% of everything that goes by.
Thanks stowe. I’ll check out both
Check out http://bit.ly/m060C where I riff on streamrolling a bit more.
130 comments!If you Bit.ly everything you, read… then you could blogroll it via a neat javascript or serverside widget.
Here is my hackjob of an scrobble experiment. It’s amazing what you can get done while waiting for your car to get fixed.A Greasemonkey script that will log all pages you visit that have a feed associated. That should cover most blogs, news etc. WSJ doesn’t play nice (but that’s for later). You can’t use a list of known blog domains because of installs on personal domains (avc.com for example).You can see what I browsed by going to http://www.ironmonkeyventur…Anyone interested in signing up can go to http://www.ironmonkeyventur…PS. If you don’t use Greasemonkey, I can zip up a Firefox extension too.
SweetI’m thinking of running a one to two month bakeoff of all of these
I use this mindmap as a blogroll http://www.labnol.org/websi…
you should try condron.us. ๐
That would be a major improvement but honestly the thing I am rebellingagainst is the tyranny of the listI don’t want to keep a list of blogs I likeThat’s creating a subset of the web and the thing I love about the web isits enormity