Make feedhaus Your Own

The feedhaus team here at bdg is very pleased to announce that we’ve added personalization to feedhaus!

The new section of the site — called “my feedhaus” — gives you the ability to organize the news as you see fit. Choose from the hundreds of tags available to create as many of your own “mini” tag clouds as you like and then drag and drop them to organize the page the way you like it.

Like the main tag cloud, your personal tag clouds are also “live” so that as the news changes, you’ll always be the first to know.

feedhaus Now Supported by Ads

You may notice that the detail pages are now ad-supported. I would have implemented this weeks ago, but Google was very slow to approve our AdSense account.

We’re not trying to make a killing here — in fact our first goal is just to cover the cost of hosting. . . .

We also implemented Google Analytics so we can track site usage.

Right now the ads aren’t targeting themselves correctly — apparently that takes 48 hours to kick in, so stay tuned.

feedhaus Now on New Hardware

In response to some complaints from my hosting company and registrar that we were using too much CPU and bandwidth, I’ve moved feedhaus from a virtual dedicated server to a dedicated server.

I think Lucene was the culprit — as FUD was indexing feeds’ articles, Lucene was consuming WAY too much CPU. Anyway, after a 24 FUDless hours, FUD is happily chugging away to bring you new stories from the 70+ feeds we now have in feedhaus.

I’m very happy to say that I was able to move feedhaus and FUD over to the new hardware with only 30 minutes of web site downtime. W00t!

This incident led me to thinking a bit about scaling feedhaus. My fears about the scalability of cometd/long-polling are probably unwarranted. What I should really worry about is FUD. I think ultimately FUD needs to be separated out from the web server machines so that it doesn’t interfere with web site performance. Furthermore, I think the feed table should probably be broken into segments and there should be a new FUD process instantiated for every 100 (or so) feeds.

I guess FUD and I will cross that bridge when we get to it . . . but for now, performance is snappy as ever AND you’re getting your news in near-realtime. Enjoy!

Podcast: Interview with feedhaus Developer Andrew Bays

The podcast returns (after another six month hiatus) with an interview of feedhaus developer Andrew Bays. If you’re interested in the inner workings of feedhaus or in finding out what’s coming soon, I highly recommend that you give it a listen.

Here are some notes from the show, which include links to most of the topics we covered. Andrew started by describing his background. He began his programming career as a volunteer developer on the Tsunami Virtual World MUD. (I commented that I was really into a MUD back in the day, the Hitchhiker’s Guide to the Galaxy Adventure Game.)

We started talking a bit about Web 2.0 and Enterprise 2.0 and how this had led bdg to conceive and develop a social news site called feedhaus. He described some nuts and bolts stuff, including how he leveraged Dojo and Comet to build the real-time features in feedhaus. (As an aside, Andrew mentioned how one of our west coast developers, Brendan Budine, ran into Alex Russell at BEA World. Alex was really excited to hear about our use of Comet in feedhaus!)

We talked about other companies who use Comet (or Comet-like functionality involving long-polling or continuations) and meebo came up in the discussion. We continued by going into a discussion about some of the myriad other technologies that make feedhaus possible. Here’s a partial list:

We closed the conversation by talking about tag clouds and the feedhaus implementation of the history slider, which was inspired by the Presidential Speeches Tag Cloud. This reminded me of another tangentially-related project, the Internet Archive’s Wayback Machine.

Enjoy the show and post your comments here or send an e-mail to [email protected].

More on the New feedhaus Features

Nearly every new user who gets on feedhaus has the same question: can I narrow the tag cloud down to only the topics/tags that I care about? My answer is no, not yet, but soon! How soon this happens is anyone’s guess, but let me update you on the progress of our newest features:

1. Search

This feature made the top of our list because you really can’t have personalization or profiling without first opening up access to all the tags in feedhaus. (In case you haven’t already noticed, only the top 100 are showing on the home page.) We’re designing search so that you can easily find feeds, articles (current or past), tags and users. The search is powered by a great, turnkey opensource search engine called Lucene.

2. Personalization

This has been the most requested feature to date. And it’ll be here . . . soon! You’ll be able to create as many personal tag clouds as you want and drag the tags you want to watch into each tag cloud. You’ll be able to name these personalized tag clouds and share them with your friends, essentially allowing you to design your own constantly up-to-date news site.

3. User Profiling and Social Networking

The user profile feature will allow you to see what other people are tagging and what feeds they’re adding (if they choose to share). You can also browse other people’s tag clouds and copy them, adding your own tags or removing those you don’t care about. If you like another person’s “taste” in the news, you can add them as a friend. Lastly, we all love a competition, so we’re adding a “top contributor” page and showing each person’s rank, which will be calculated based on how many feeds he or she has added.

Be on the lookout for these new features and, as always, leave your comments here or drop us a line at [email protected].

feedhaus Alpha 2 is Live

We made a few notable changes, including revamping the “add a feed” page by adding clearer feedback and better instructions. We also made the feed adding, indexing and aggregating a lot more robust by fixing some bugs deep within the feed processing engine.

We’re trying our best to emphasize that this is a social news site, so we added the orange button (on the right) to encourage people to add their own content.

Coming in the next alpha build: the much anticipated search feature, along with perhaps some personalization and/or user profiling bits. More about these new features to come….

feedhaus Public Alpha Begins . . . Now!

feedhaus_public_alphaWe’ve put the “Alpha 1” build up and removed the password protection, which marks the beginning of the feedhaus public alpha!

In this build, we’ve repaired a lot of the IE problems (although there are still a few sneaky issues) and revamped the forms along with other parts of the UI. The “add a feed” page is much more robust now — if you enter a web site URL instead of a feed URL, it will actually search the page for one or more feed URL(s) and pre-populate the form for you. (Thanks Andrew for implementing this great feature.)

We’ve also changed our slogan from “What’s Hot Now” to “Be the First to Know.”

One known issues is that the history slider is still a little wonky and it will be for several days (until enough snapshots exist for it to scroll smoothly). This problem will be with us for about a week or so and then it will work itself out. Think of it as feedhaus’ “growing pains.”

So, have at it, folks!

And please report problems and suggestions by commenting on this blog or sending an e-mail to [email protected].

Web 2.0 and Cascading SLAs

twitter-down-againHerein lies a great example of one of the downsides of Web 2.0. It’s something I call “cascading SLAs.”

I was checking feedhaus today and I noticed that the flickr badge (the cool lil’ widget that displays a little collage of photos) was down for the count. I thought perhaps I had broken something on the dev server, but a quick health check revealed that everything was okay with feedhaus.

So, I decided to check out flickr to see if there were any messages about known downtime, current server issues, etc. Lo and behold, flickr was also down! Hello? Anyone? Bueller?

This demonstrates one of the classic problems with mashups, a crucial component of Web 2.0: cascading SLAs (Service-Level Agreements), or, more precisely, a lack thereof.

Here at feedhaus, I have a responsibility to provide up-to-date news so that my users will be the “first to know.” I can (although I probably won’t) guarantee a level of service for feedhaus’s ability to deliver content. But, as a multi-band content aggregator, I’m solely dependent on the sources of content — namely flickr, YouTube and you-name-it syndicated feed from whatever.com. Now if my sources are CNN, Google, Fox, etc. I would expect pretty dependable service. But Digg? Twitter? Seeing what happened to Skype recently, I’m beginning to wonder if everyone, even the biggest — and most distributed — systems are subject to serious unplanned downtime.

So, what I’m getting at here is that my SLA, no matter how much I pay my attorneys to draft it, is only as good as the SLAs of the services that I use. Now, I’m not paying the sources for those services — and, I might add, you’re not paying me to use feedhaus — so I have no SLAs for my underlying services, which makes my SLA worth less than the paper it’s printed on. Do you see the problem? (I’m reminded of a certain scene from the 1989 classic comedy Major League. “See, it says right there; no calisthenics. What do you think of that?”)

So, before you start drafting that SLA for the cool new mashup you just built between Google Maps and Facebook, think about the stability and sustainability of your sources. Or else your SLA might have the same fate as Roger Dorn’s contract. . . .