If Anyone Wants to Give Me Some Feedster Comments ...

Here are two urls to try:

• Search #1
• Search #2

Tell me which you like better.  I prefer #2 and I think you will also.  Tip: Open them in two separate windows vertically side by side.  This isn't production code yet so the search #2 could break.  It shouldn't before tomorrow though.  Leave me a comment.

Night.  Not much in the morning -- meetings until the afternoon.

Oh and this was really cool.  Go Ad Rants!

10:25:49 PM      Google It!   comment []    IM Me About This   

Feedster Nightly Update / It Takes a Village / Miscellaneous Thoughts

Oh and our additional server capacity will be up and running with 72 to 96 hours.  That will enable some of the more intensive things we want to do as well as make everything more reliable in general. 

Please note that we may be offline Tuesday night from Midnight EST for a bit.
If this changes I'll let you know.

We've all heard it said "It takes a village to raise a child" and perhaps in these complex times, "it takes a village to make a product".  I'm getting great feedback from all over the globe.  Thank you to everyone.  Examples:

• Over in Greece, I got good thoughts on scheduling for the "crawler" (I don't think of it as that since its not recursive (well kinda but ..))
• From Holland, came good thoughts on finding more RSS urls and other important things
• And Andy Edmonds tonight helped me tune the query parser for *much better* search results.  Thanks !

Not a ton of new features to report except for the tuned query parser.  A lot of work on back end stuff and getting ready for more server capacity.  A cool new feature, very appropriate for the times, is coming in a few days.  More when I'm comfortable with its capabilities.

Michael Bernstein sent a few great links for the Wrox debacle: 1, 2, 3, 4, 5.  Now when he was doing his research, he found that Google was indexing Feedster pages!!!!  See the 4th link from the bottom on this Google page.  How'd that happen ?

Note: You'll notice I didn't use my convention of [_Go_] for the links above.  Here's why -- we're starting to add link anchor text analysis in and clearly if I know "Andy Edmonds" is related to urls at mozilla / surfmind.com, we're going to do a better job.

Marc's Open Indentity is very, very cool. 

Via Dave: Great article on Google.  Now that I'm back in the search engine business, I'm definitely reading this one.  Back in the search business you say.

ClickTracks: This looks like a very useful web analytics tool I just ran across.

Orange Alert from Chris Pirillo.

Oh and did I mention that we're up to almost 80,000 posts in the database ?  Very cool.

8:55:34 PM      Google It!   comment []    IM Me About This   

Feedster Gets a Bit More Attention

I was just interviewed about Feedster by an Infoworld reporter.  Interesting.  Now I'm not praising this by any means but we've put less* marketing effort into Feedster than I have for any other product in my career.  That's a testimony to the power of blogs if I ever saw one.

*that could also be written as "no marketing effort at all".

5:36:46 PM      Google It!   comment []    IM Me About This   

Anyone Know Anything About Wrox in the US ?

A lto of us technical types have books from Wrox (those are the red books).  Apparently in the UK they're in trouble:

It would appear that Wrox press have gone into insolvency. Their investors pulled the plug. All Wrox employees are now out of a job, whilst the recievers work out what's saleable and what is not. All accounts payable are on hold -- no-one will get paid till this is resolved. [_Go_] and [_Go_]

Now the question is "What about Wrox in the US?".  Comments anyone?

Oh and a big thank you to James for some MySQL help earlier today.  Much appreciated. 

3:12:31 PM      Google It!   comment []    IM Me About This   

How to Add Feedster to Mozilla and Other Cool Mozilla Stuff

From "Andy a true MacDaddy of Mozilla" (at least from my biased perspective) comes how to add Feedster as a search option right to Mozilla.  

Feedstering Mozilla

Time to set up a new Mozilla keyword shortcut RSS.  A few steps in the process: [_Go_]

.. (steps to do this)

Screenshot of the resulting dialog.  Now, you can type "rss foo" in the url bar to search for foo on feedster.

Very nice.  And if you're into Mozilla (heck even if you're not) then you should check out:

• Junkie
• Blozom
• The GoogleBar

1:18:57 PM      Google It!   comment []    IM Me About This   

Feedster

Hi, just a quick note.  I'm seeing performance that is off a bit today and we're being hammered with hits.  Have a litle patience.  I'm ordering more servers now.

11:37:50 AM      Google It!   comment []    IM Me About This   

KDE and Feedster

Although I mentioned this last week, now that made KDE News, its even more official (although is that more definitive than a CVS archive?  Depends on your point of view I guess):

You already have KNewsTicker and you often hit the popular sites for the latest news. But have you tried Feedster, the RSS search engine? Now, the latest KDE CVS includes support for Feedster in the elegant form of a web shortcut. [_Go_]

10:34:29 AM      Google It!   comment []    IM Me About This   

Nicest Comment of the Day

You just made my day:

Sometimes someone has a brilliant idea, making the Internet take a step forward. [_Go_]

Yes he's talking about Feedster.

8:41:55 AM      Google It!   comment []    IM Me About This   

Internationalization and Translation Issues

I'm getting a huge push on this now.  And I understand the demand / need for it.  Trust me we're working on it.  If you want to pitch in on the user interface translation, drop me an email and I'll be in touch when we're ready to start on it.  I'm working on getting language categorization in very quickly results NOT in your native language can be filtered out of the result set.  And proper indexing, etc.

8:29:08 AM      Google It!   comment []    IM Me About This   

And Here's What the Community Did

Well not only was I busy this weekend but others were too.  Here are some of the highlights:

The FurryGoat (I love that name) tells you how to add Feedster to Internet Explorer's search options using TweakUI.  [_Go_]

Feedster is increasingly being added as an option right next to the "Google It" links on blogs.  That's high praise.  Here are some examples:

• FurryGoat
• BunkoSquad (thanks Sooz!)
• Vicky (a cool multi search engine interface)
• (If you're adding Feedster to your blog like this, drop us a line and we'll link to you / add you to a list of "Blogs Supporting Feedster" that's coming out soon)

Ben Meadowcraft of the Mozilla Mycroft project is working on adding Feedster support to Mozilla's ability to drive Apple's Sherlock.  I'm not 100% sure on the status of this since I haven't personally tested it but if you want the .src and .png files for it drop me a line.  If Ben says its ok, I'll link to them as well.

The Amazing US contributed a really nice CSS design for our results that while we haven't had time to implement yet is coming.  Thanks!

I'm sure there's more going that I'm missing but this is a blog.  There's always more!

8:26:40 AM      Google It!   comment []    IM Me About This   

It's a Beautiful Morning in the Feed(ster) Neighborhood

Wow.  It was a *busy* weekend.  Here are the latest changes to Feedster we made over the weekend:

• The database has grown to 73,331 posts, 5,302 RSS urls and 87,000 plus queries.  Wow.
• With help from Sjoerd, we fixed the problem with Feedster's RSS feeds showing up in Radio's aggregator.  Thanks! We've now tested them in Radio, Ampheta and Net News Wire and if there are any other problems, let us know.  [_Go_]
• Thanks to a gentle nudge from Chris Pirillo, we added (trumpets please!) sorting search results by date!  This is a big improvement since it helps you (a lot) with what's going on now.  Oh and if you didn't know, Chris was recently chosen as the best Computer / Technology Blog.  Congrats.
• Improved our suggestion / problem code with the ability to respond directly to inquiries from the inquiry, not a separate email program.  That's minor but it'll help keep us responsive.  And I'm a huge believer in customer service.  Even in this context of a free service.
• Added some additional intelligence to the result ranking so for some terms (and its not huge yet), the "definitive" result bubbles to the top.  Take this search for example.  You aren't going to find that keyword anywhere in any of those postings.  Why?  Because he doesn't talk about himself.  Still that's the "definitive" url for this community.  More examples: 1, 2, 3, 4, 5, 6,7.
• Brought back two features we had (briefly):
• Filter or the icon strips out of the search results any results for that blog.  Example -- you want to search for something but you don't want any results from your own blog to show up.  Just click on the  icon next to your blog in the results list.  This wasn't my idea but my brain's too fuzzy right now to cite the right person.  Sorry.
• Drill or the  icon shows you all the posts we have for the selected blog.  Very useful when you know its there but just can't think of the keywords.  You could also think of this as a "Virtual TOC" for a weblog.
• Added support for the Apple Sherlock recommended "search comments" that let Sherlock know how to drive other search engines.
• Attempted (un-successfully) to fix our style sheet for Internet Explorer on the mac.  Looks like its back to browser detection and tables for that platform.  Ick.  (also look for a css rant later today)
• A bunch of enhancements to the core data structures and content processing / loading.  Doesn't mean much to you I know, but it means more good and frothy features will be flowing your way.
• And I think there's more but I'd have to look hard for it to figure it out.

Overall?  The service just works better.  Let us know if you're happy (or not).

 

7:40:50 AM      Google It!   comment []    IM Me About This