Follow Up: Feedback Needed

Posted by Bill @ 12:29 am

If you're new here, you may want to subscribe to my RSS feed. Thanks for visiting!

Thanks for the feedback both on the board and via email. Great stuff.

Part 1: I’ve created a mailing list called lapoker at goodtimeslist.com. You can send a subscribe request to lapoker-subscribe. It’s a subscription moderated list which means that I’ll have to approve you before you’re added to the list but once on the list it’s unmoderated.

Part 2: Some interesting emails on this one. First, I’ll try to explain it in a little more detail as there seemed to be some confusion. This wouldn’t be a replacement for Bloglines, GatorNews, SharpReader or any other RSS/XML reader. In general terms, what I was thinking of is that instead of going to example.com to get their RSS feed, it would be spidered by the aggregator program and all of the story links back to example.com would be modified so when they were clicked on it could keep score as to what was being read.

The idea is that you’ll end up with two types of scores; the first being for an entire site and the second being for individual posts. So, let’s say in a universe of 200 poker blogs, all 200 are being read by the people who use the aggregator feeds instead of getting them directly from the source. I certainly don’t have time to read 200 blogs every day but there’s certainly some quality postings I might be missing out on, on the sites I don’t read regularly (or at all). But if a post on a blog I don’t read gets enough clicks from other people, it would be included in a list of today’s top 50 (or 100 or ??) posts of the day and I can subscribe to that XML feed and see the day’s most popular blog postings.

Obviously, a blog like Iggy’s would always rank very high because he’s likely to have many, many people subscribing to his feed. I’ll have to come up with some sort of scoring model to even things like that out a bit but hopefully the result is a good mix of quality and popularity that eases the data burden on people trying to keep up with what’s going on in the poker blogosphere.

Based on the responses it seems to be something with some demand so I’ll keep kicking around some ideas on this and maybe work up a small prototype fairly soon. Unfortunately, my schedule is a little hectic the next couple of weeks. I’m in NY and NJ next week and then up in Portland for a week in the beginning of March. I’ll try to squeeze some time in here or there to see what develops.

If anybody has any ideas or wants to lend some PHP or Perl skills to this, drop me a line.

License

This work is published under a Creative Commons Attribution-NonCommercial-NoDerivs 2.5 License.

If you enjoyed this post, please consider getting our free RSS feed so you can be notified of other posts like this.


COMMENTS / 7 COMMENTS

I’m always around and can help with PHP stuff if needed.

Ryan added these pithy words on Feb 15 05 at 4:52 am

i’d also be happy to help with any php or perl geekiness.

fhwrdh added these pithy words on Feb 15 05 at 7:50 am

Is RSS really the right technology to use for this, Bill?

(For everything that follows, keep in mind that I’ve only dabbled in RSS, having written a few screen-scraper type syndicators of sites I was interested in that didn’t offer their own feeds)

At first blush it seems like a good fit, but it seems that if you’re trying to do something close to real-time rankings you’re going to have a “new” (which would mostly just mean slightly reordered) feed every couple of minutes, which means that every aggregator that pulls down every ten minutes will always be getting a new feed from you… I hope you’re not planning on including the entire post in each entry? Or you have a sufficiently fat pipe?

Also, some RSS readers (NetNewsWire for the Mac leaps to mind) won’t necessarily reorder if the item (title/link/whatever) stays the same even if its position in the RSS feed changes. That might be contrary to spec, though. This also might not be a problem if you’re offering the “top X” as a group rather than an ordered list.

Of course, having said that, a) I might be missing something in the 2.0 spec that handles just such cases as this; and b) it sounds like fun! Sign me up, if you actually need more help than has already been offered.

Tapin added these pithy words on Feb 15 05 at 8:51 am

> Is RSS really the right technology to use for this, Bill?
>

I’m always open to alternate solutions. I’ve gotten pretty far in tech by admitting I don’t know everything :-)
> (For everything that follows, keep in mind that I’ve only dabbled in RSS,
> having written a few screen-scraper type syndicators of sites I was
> interested in that didn’t offer their own feeds)
>
> At first blush it seems like a good fit, but it seems that if you’re
> trying to do something close to real-time rankings you’re going to have a
> “new” (which would mostly just mean slightly reordered) feed every couple
> of minutes, which means that every aggregator that pulls down every ten
> minutes will always be getting a new feed from you… I hope you’re not
> planning on including the entire post in each entry? Or you have a
> sufficiently fat pipe?

Good point. I was planning more on abstracts rather than entire posts. Maybe first 50 words or so.

Also, the XML/RSS feed would not be dynamic (at least that wasn’t my thinking). It would publish a static page once an hour or so (via cron). So people who are checking every 10 minutes will get a “Not changed” and shouldn’t re-fetch the page (if they actually respect the http protocol). If push comes to shove you can always throttle connections with an Apache module to keep the bandwidth suckers from causing too much grief.

>
> Also, some RSS readers (NetNewsWire for the Mac leaps to mind) won’t
> necessarily reorder if the item (title/link/whatever) stays the same even
> if its position in the RSS feed changes. That might be contrary to spec,
> though. This also might not be a problem if you’re offering the “top X”
> as a group rather than an ordered list.

In regards to my previous statement above, I view this more of an annoyance than anything else. Also, not all RSS readers handle this in a consistent manner so . . . not much I can do about that.

>
> Of course, having said that, a) I might be missing something in the 2.0
> spec that handles just such cases as this; and b) it sounds like fun!
> Sign me up, if you actually need more help than has already been offered.
>

I’ll keep you on the list! Thanks!

Bill Rini added these pithy words on Feb 15 05 at 9:21 am

I’ve written a few RSS readers in perl and could whip this out in a few hours by copying old code and combining with a survey program I wrote a while back. All it will cost you is a link to my stupid blog on your site.

Bob
thedegenerategambler.com

Bob added these pithy words on Feb 15 05 at 5:12 pm

Hi there old chap. I am just enquiring to see whether or not your impressive computer literacy skills are also present in the field of hay farming? Me and my friends have been contemplating growing some hay for fun. You know, for the summer time! Anyhow, we were wondering if you could inform us of any recent, vast hay growth you have encountered recently? Thank you in advance for any wisdom you can cast over this great shadow of confusion we are currently in. Reply to we_love_hay@hotmail.com.

Yours,
Dave and the Farmer Boys.
xxx

Dave added these pithy words on Feb 15 05 at 7:59 pm

So if I’m reading it correctly, you’re NOT going to allow users to be able to re-fetch pages on demand, correct? I wouldn’t want you to open yourself to hackers trying to bring things down using the re-fetch in a loop..

BTW, I’m now PHP savvy :D
I see you already got a bunch of people that said they’d help, but I’m just letting you know I’m here in the case you need ;)
Bloggers r a geekie bunch, eh? -grin-

Wait.. can I put in more net-speak?
Hmmm, ah, next time :p

~Glenn

Glenn added these pithy words on Feb 17 05 at 10:44 am

SPEAK / ADD YOUR COMMENT
Comments are moderated.

XHTML: You can use these tags: <a href="" title="" rel=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <card> <code> <em> <i> <span class=""> <strike> <strong>

Return to Top