Posted in projects, technology by johnsnavely on October 15, 2008

There’s an interesting dicussion going on in the comments, that I think should be elevated to a post.

Basically, it revolves around Google Reader, a nifty web app T introduced me to a couple years ago, which basically changed my life. I use it everyday.

The question (for you Reader users out there) is this: If you were to redesign Google Reader? What would you add/remove? How would you change it?

More importantly how do you use it? I’ve been using it to filter all my status updates from flickr and facebook, and as a shopping “bot” on craigslist and ebay (that’s how I found my current apartment)


One feature I would offer is the ability to repost an item from reader as a blog post using google sites. I’m not exactly enthusiastic about encouraging recycling, but it’s basically what we do a lot of now…


13 Responses

Subscribe to comments with RSS.

  1. son1 said, on October 15, 2008 at 9:02 pm

    I think reposting is (somehow) a great idea. Of course, one could think of delicious as a style of reposting — a style which permits a 1000-character edit.

    The danger to most of the suggestions in the other comment thread must be the inevitable feature drift. Google Reader’s strength (or really, any web app’s strength) is its simplicity: lists of posts, read and not read, shared and not-shared, starred and not-starred.

    Anything that goes beyond that probably needs some fairly heavy-duty justification.

    Anyway, a related question: what if you wanted to build, not a system for post-processing and filtering items in RSS feeds, but a system for building such a system? What if we were thinking about Google Reader in an infrastructural way? Can we imagine building a platform like this, something programmable, that doesn’t (in the end) come out looking like Pipes?

  2. son1 said, on October 15, 2008 at 9:26 pm

    Just saw this.

    But the Harvard economist [Dani Rodrik] finds the blog — short for Web log — useful because it serves as a reference catalog for his ideas. “I now constantly Google my own blog for ideas that I knew I had at some point,” he says.

    Which is just a way of routing around the Web’s own un-useability.

  3. johnsnavely said, on October 16, 2008 at 1:21 am

    Cool. Thanks for the link. I delicious’ed a couple links I saw on Swiss MIss that inadvertently dealt with reader.

    Couple o’ points. I agree that Reader does what it does simply, which is one reason for its success. But, like Bryan points out, right now, it’s a big fat cul de sac, when it has the glimmer be “something else”, something of a lifestyle shift.

    What I’m mostly interested in use. For me, as a I said before, reader might have the potential to become my communications hub. (At one point, I was trying to get my google calendar to spit out events into an rss so I could “read” my calendar in Reader. Recasting reader with a different use doesn’t have to expand it to some feature bloated useless thing. But what if you used your reader as a calendar or vice versa. That’s why I also asked about features that you would drop.

    Much of this thought actually stemmed from a realization I had a couple weeks ago. I hadn’t been using reader much, and I realized that reader is important to me, but it’s not connected to my life. I could probably give it up.

    But delicious has been “part of me” for a longer time, and it’s not going anywhere. If delicious wrote their own reader as a firefox plugin, wouldn’t you use that instead of google’s? delicious “subscriptions” are only halfway there.

  4. johnsnavely said, on October 16, 2008 at 1:43 am

    So onto your second question, the “layers of abstraction” question.

    I’m a little worried about this question. At some point, we keep abstracting out until we have to make our own browser, which is lame, according to some people.

    That caveat aside, the tools to build my own personal version of reader would probably look a lot like pipes. I’d have a place to collect things, maybe an infinite canvas. I’d have different ways of visualizing (reading), organizing, and sharing my collection. I’d have little widgets that I could use as filters or write code snippets into to place on top of visualizations or collections of object. And I’d have an easy “save as” function that would allow me to pipe my collections elsewhere. To design your own reader, you pick a collection box, a visualization style, and a set of pipes in and out…. Or something, I’m writing this back of envelope so I don’t know…

    What were you thinking?

  5. son1 said, on October 16, 2008 at 1:51 pm

    What if it was simpler. What if it was just “Reader, but with plugins.” You could write little Javascript modules that could be plugged-in as preprocessors, or postprocessors, of feed streams. And maybe the Reader had a bit more of a flexible layout architecture, which could be programmed to by the plugins.

    So “by default” the reader would collect all your RSS streams, show you only the new items, and show them to you in a time-ordered list.

    But then you could write plugins where … some items would get extra colors. Some would pop up, not in the list, but in a tree-structured hierarchy. And other plugins would see those entries, and then go off and do things like, “fetch all of the posts from my blog with some similar words, and display them next to the original post from the RSS feed.”

  6. son1 said, on October 16, 2008 at 1:58 pm

    Much of this thought actually stemmed from a realization I had a couple weeks ago. I hadn’t been using reader much, and I realized that reader is important to me, but it’s not connected to my life. I could probably give it up.

    What other feed-readers have you looked at? Have you considered moving to a Firefox-based reader?

  7. johnsnavely said, on October 17, 2008 at 1:11 am

    I like the plug-in idea a lot.

    But first, I think I need to spend some time with other readers. That xoogler link I sent has a list of people from reader who have moved on to other reader-like projects, like friendfeed, which is actually closer to what I’d imagine reader being like.

    Once I got this new computer, I switched up a few things trying to streamline how I work… I realized how little experimentation I do and how many things I’d like to have are already out there.

  8. bryan said, on October 17, 2008 at 1:42 pm

    plugins would indeed be excellent. even something simple like, “I want to subscribe to this feed but only show me posts that mention X and Y.” I’m sure there are other newsreaders out there that provide this functionality but…. I dont want any more desktop apps.*

    An aside: Have you guys played with yet? I love the clarity of it, but I wish it allowed inputs via RSS or some other system.

    In general, I keep waiting for a new class of webapps that are more like single function widgets. If I could get my feed management from google, my visualization from daytum, my filtering from somewhere else, etc I could encrust the services and functions that I care about around the feeds that matter to me most. All these years and the web is still dominated by sites striving to dominate. That is, sites that want to be a destination. Perhaps this is the mythical world of yahoo Pipes? Do you guys have some recommended reading for pipes… I still haven’t looked into it at all.

    About abstraction, I think there is a position which is between abstraction in the way it’s being used here and and the discrete interface of an application. For instance, the gmail search filter and itunes “smart playlist” creator are middle-road options: they allow the user to add a minimal feature to the app. In both cases you are able to build fairly sophisticated queries in a manner that is user friendly and yet it’s somehow avoids the ugliness that is a wizard. Maybe because all the options are in one window?

    It also occurs to me that I would like to see a reader that is smart enough to group posts by topic. A sort of smart digest. If 3 different blogs are talking about Madmen, the reader should allow me to follow that cross-blog thread by exposing a temporary channel. Bonus points if the software is smart enough to recognize duplicate or similar text and the re-posting of images.

    Personal assistants are the new feed reader.

    *I do really wish they would port Google Gears to the mac!

  9. son1 said, on October 17, 2008 at 4:50 pm

    John, we need to get bryan involved in some of these offline discussions we’ve been having…

  10. bryan said, on October 17, 2008 at 6:07 pm

    As long as I get to keep my kidneys…

  11. johnsnavely said, on October 17, 2008 at 7:54 pm

    You can keep one.

  12. bryan said, on October 19, 2008 at 5:13 pm

    OK, I played with pipes a bit and now I see why you guys like it so much.

    Help the noob though, I am sort of stuck. What is the work around for getting a item.attribute into string format so you can use it as input on other modules? That this recasting has been omitted from the system is *so* aggravating.

  13. RE: Re-Reader « John Snavely’s Blog said, on October 24, 2008 at 4:37 pm

    […] I had posted before about my frustrations and the many flaws of google Reader. (I was not alone.) And lo! My prayers were answered with just a short search to find other […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: