onfocus

A conversation from around the house earlier this week...

pb: (excited!) did you see the latest post on my site?
shawnde: Does it have anything to do with Amazon?
pb: yeah!
shawnde: (rolls eyes.)

Daschle will post diary on his Senate Web site: "'My staff told me a while back about a blog,' Daschle said. 'I learned a new word. ... This new blog concept appealed to me.'" [via mkelley + blogpopuli]

Several people have mentioned that it would be nice to show the newest products in the Amazon RSS feeds rather than the top-selling products. There's a quick hack to make this happen. Build a feed with the Amazon RSS Feed Builder like normal, click "get feed", and note the local onfocus.com URL for the resulting page in your browser's address bar. The last bit of the URL should be %2Bsalesrank. Highlight this text, change it to +daterank, and hit enter. When the page reloads, you'll have your RSS feed URL sorted by product release date.

Note that the RSS feed description will still say "Up-to-date listings of Amazon.com's top items (by Sales Rank)", even though it's sorted by date. This is a function of Amazon's stylesheet that's transforming the AWS response—this phrase is hard-coded.

Kinda kludgy, but it works! I set up some feeds to watch for new books by publishers I like. I'll see if I can work sorting choices into the interface when I get some time.

I didn't sign up for this. FTLOG, I had no idea Oregon would be this hot consistently. Luckily it looks like it's going to cool back down to the 80s in few days.

More Amazon news: According to the Seattle paper it turns out Ted was not a PR-experiment fiction, but is a real life person. The only question that remains is why they yanked the plug on his blog so abruptly.

Amazon didn't really let people know how to add feeds to their newsreader so I whipped up a page that can get you started: Amazon RSS Feed Builder. You can build Book, DVD, or CD feeds with this little tool. Just choose a category or enter a keyword and you'll get the URL for the feed. Then you can copy the feed URL and paste it into your newsreader. If the RSS looks good (ie. has items), you should also get a preview of the feed so you can see what's in it. The feed builder hasn't been tested too much so let me know if you run into problems.

Here are some RSS feed URLs I built with this tool (that I'll be monitoring):

• Top Computers/Internet Books
• O'Reilly Hacks Series
• Top Documentary DVDs
• Top Electronic Music
• Top Bargain Books

Try it out: Amazon RSS Feed Builder.

Update: The feed preview is a bit sketchy—so even if it says there aren't any items, check the feed manually to see if that's true.

Matt notes that Amazon has embraced RSS syndication of its catalog. Matt also posted an example feed that lists books about weblogs. This is a great illustration of why Web Services are important.

Speed is key

To implement this new feature in the world before they had Web Services, a company like Amazon would have to focus some developers on putting together the RSS. This would involve special code to get the results from their database and "hard coding" those results as RSS. Any change in use would probably require new code, more developer time, etc. In the world with Amazon Web Services (AWS) though, they already have output in an XML format and a mechanism for transforming that XML (XSLT). All they needed to do to implement this feature was write a stylesheet to transform AWS responses into RSS. Instead of writing new code to accomplish a specific task, they can simply write a new stylesheet to fit their new task. And the stylesheet will work for almost all of the AWS methods available (Author Search, Director Search, Keyword Search, Category Listing [aka. Browse Node]...on and on). With a single stylesheet they've opened up their catalog to the thousands of people using RSS newsreaders.

So, how does it work? Check out the monster URL for the sample feed Matt posted:

http://xml.amazon.com/onca/xml3?t=webservices-20&dev-t=amznRss&KeywordSearch=weblog&mode=books&bcm=&type=lite&page=1&ct=text/xml&sort=+salesrank&f=http://xml.amazon.com/xsl/xml-rss091.xsl

The URL is just a standard public AWS query with a slight change. Instead of the normal f=xml in the URL, it's f=http://xml.amazon.com/xsl/xml-rss091.xsl. This tells AWS to use Amazon's RSS stylesheet to transform the results before sending them back. And once you see that it's using their standard AWS queries, you can use that to your advantage.

What's in it for me?

If you want to offer specific Amazon product feeds, and you want to get a kickback (ok, affiliate fee) when someone clicks the link through their newsreader and buys a product—just tweak this URL. Change t=webservices-20 to t=[your associates tag] and dev-t=amznRss to dev-t=[your developer key]. That's it! Nothing to install on your server, no code to write. The only work you have to do is encouraging people to add your URL with your variables to their newsreader. This economic incentive is how Amazon encourages people to use their system, and I'm amazed more companies aren't doing it.

As Matt pointed out, Amazon will make money with RSS. And thanks to their open platform—there's no reason you can't make some money too.

There's an article in the Oregonian about the upcoming Blogathon. I won't link to the online version of the article because the Oregonian has a severe usability problem. Every time I click on a "deep linked" article, I get hit with a splash page that says, "Help Us Serve You Better" and asks for my zip code, age, and gender.


OregonLive serving us all better

Saying that it "serves me better" doesn't seem to be accurate, though, because the content is not personalized to the information I put in. (I'm frequently an elderly woman from Ohio.) I don't think they're actually serving me better, I don't think they care about my experience at all—they're simply looking for demographic information. (Most likely so they can tell their Web advertisers who they'll be reaching—hmm, lots of elderly women from Ohio!) And in the process, they're getting in the way of how I want to use their site. To make it worse, once I enter the info I go to yet another splash page that is filled with disclaimers and legal warnings. (Tip: Never let your lawyers welcome people to your site.) By the third click I finally get the article I wanted to see—and it's covered with screaming, zooming, flash, animated advertising. Of course, most of the time I don't click through all of these pages. (And consequently, I don't see their ads. Which means they're loosing money.) Why don't newspaper sites get the Web? Anyway, if someone mirrors the Oregonian article I'll link to it. But I don't want you to go through the clicking hell.

Instead, search for "blogathon" at Google or Daypop and see what people who understand the Web are saying about it. Don't get me wrong, I'm glad blogathon is getting mainstream press. People outside the weblog community should know about it. It's a fun event for some very good causes, so definitely check it out. (And look into sponsoring some poor, tired, 24-hour blogging people if you have the means.)

Life in Roratonga: "When we tell our new landlord that we'd like to move into her house, she slowly gets up from the garden she's tending to and asks us what day it is. Then she asks which month."

O'Reilly put up some beta Amazon hacks from the book of the same name. These hacks haven't been through the entire editing process, so they're a little rough around the edges (hence, beta). Each hack (all separate PDFs) also has a complete table of contents, so you can get an idea of what's in the whole book. (Did I mention you can pre-order on Amazon? ;)

mistimed jump

An indpendent Amazon developer is keeping his eyes on the developer discussion board at Amazon and highlighting the good stuff on his site. [via the AWS Newsletter] This is cool, but it would be even better as a frequently updated weblog.

weblogs.com hasn't updated since 11:45 this morning. Which means Weblog Bookwatch and ORblogs aren't updating properly. (And a bunch of other services around the Web, I bet.)

Meg has a great idea for revitalizing tourism to Niagra Falls. Another way to improve tourism would be scheduling "natural flow" days at different times of the year where they open the dam a bit and bring the water flow back to its pre-dam volume. It would be a spectacular sight to see. In fact, people are trying to restore the flow. (Though I'm not sure how actively, because that page is out of date.) They could also go the other way and have "no flow" days where they let people see the rocks underneath the falls. (They did this once for research.) Though both of these may take away from the fact that Niagra Falls is a "natural" wonder because people would see firsthand that we control the amount of water rushing over the edge. Of course the least desirable option is to let Pfizer rename the falls for millions of dollars. It's only a one-letter difference—changing the signs would be easy.

If you're going to be anywhere near Portland tomorrow, be prepared to celebrate building the web at Webvisions. I'll be there—learning all I can. (Not sure if they'll have fighting robots, but I don't see why they wouldn't.) Rumor has it there will be wireless access, so I'll try to post what I'm learning.

If you're going to be anywhere near Corvallis this weekend, be prepared to celebrate art, science, and technology at da Vinci Days. I've never experienced them before, but I'm looking forward to it. The film festival looks interesting. And what small town festival would be complete without the traditional fighting robots? Or a giant robot (Juggerbot) destroying appliances? If I remember my history correctly, da Vinci had sketches of giant fighting robots but technology just wasn't ready to build them. So ahead of his time.

cool! Matt started PVRblog—news/how-tos/reviews about TiVos, ReplayTVs, and any other DVRs.

For $195 you can read the Jupiter Research report on using blog software for content management needs. Or for $0 you can read Matt Haughey's Beyond the Blog essay about using Movable Type for more advanced content management.

Dana Milbank at the Washington Post dissects the evasive tactics used by Ari Fleischer in his final meeting with the press. If you're thinking about becoming a public spokesman that has to continually evade subjects in the midst of difficult questions that it would be better not to answer, you may want to see how he does it. [via blargblog]

Ted, the Amazon blogger I posted about a few days ago turns out to be a PR experiment rather than a real person. Seattle Post Intelligencer: "'Without going into whether the content is factually accurate or not, what I can tell you is that it was not an individual's personal blog. It was a recruiting message that we put in the form of a blog to experiment,' Curry said." They've taken the fictitious blog down. Maybe "Ted" was an acronym for Targeted Enlistment Device.

I updated ORblogs a bit. The site now has a weblog called ORpost that grabs the latest posts from Oregon weblogs' RSS feeds.

Lawrence Lessig is going on vacation so he found a guest blogger to fill in while he's gone: Presidential candidate Howard Dean.

Tom Tomorrow's weblog is driving book sales. [His new book.] (He needs an associates account so he could make some extra money from those links. Which is explained in detail in another new book that will be available soon.)

It feels to me like the weblog developer (and user) community could learn a lot from the Open Source community's struggles with closed vs. open standards. This afternoon, Peter Saint-Andre gave a talk about the state of IM, the various protocols (AIM, MSN, Yahoo) with varying degrees of openness, and the emerging standard XMPP (open) protocol that he's involved with. Just because there's a standard in place doesn't mean it's necessarily developer-friendly, especially if controlled by a large commercial interest. I could imagine a talk four years from now that is essentially the same, but replacing "jabber-based protocol" with "n(echo)". In addition to an open standard, and supporting open source code, he listed an open community (with a standard process for extending/improving) involved with guiding the standard as an important requirement. It'll be interesting to see if weblog software follows the same path as IM software because there are already quite a few parallels. Will n(echo) eventually move to the IETF, IBM, Google, Six Apart, or will there always be a loose consensus guiding it? I'm not involved with that project, and maybe these sorts of questions are already answered. It just seems like there are several similarities between weblogs and IM, with a chance to learn from the recent past.

How many people are blogging? In what languages? NITLE's Blog Census has hard data. The guy who wrote this (didn't catch his name) gave several interesting bits of data, including this gem: 2% of Icelanders have a Blog*Spot blog. There's an API. They also track weblog tool usage stats.

Update: Blog Census is by Maciej Ceglowski. Thanks Anil!

Some interesting points from the conference so far today:

• Developing countries equate software and Microsoft. They don't know about alternatives.
• Bradley Kuhn noted that most of the world isn't using computers yet—which means they haven't chosen an operating system yet. He argues that free software needs to expand to developing countries.
• Technologies like VXML (Voice-XML that powers automated phone systems) are more critical in non-literate societies.
• 33% of eBay's listings come in through their API. (That's millions of listings.)
• There's an analogy between the Domain Name System and current Web Services. Users and developers need to guard against lock-in points like the DNS government-granted monopoly to NetSol.
• Tim quoting Lao-Tzu: "Losing the way of life, men rely on goodness. Losing goodness, men rely on laws." Licensing agreements change. Company's strategies change. Since we're not following The Way, how do we hold on to the current Web Services goodness?
• The business case keeps the goodness according to Amazon and eBay. Both companies feel the economic pain when their developers fail, so they want to keep the developers successful. (Unlike traditional software platforms.) [My note: Though traditional software vendors do feel the pain of developer failures in the form of lost future revenue—unless you're a monopoly. But those are illegal anyway. heh.]
• Google declined to participate in the Web Services Bill of Rights talk.

This afternoon you can probably find me at:

• Emerging Topics Lightning Talks
• Open Source and Open Standards
• Amazon Web Services: Past, Present, Future
• Transforming XML for Web and Print

whew!

I had a great afternoon at the conference. As someone who codes in isolation for days/weeks/months on end, it's nice to see powerpoint slides that show real live code by other people. Though the Filtering Email with Perl session turned into a bit of a code-critique by the audience. (Tough crowd.) It was still fun to hear a bunch of Perl hackers discussing how to best optimize the script in the presentation. In fact, they could turn that into a session: Let the OSCon Audience Optimize Your Script. You could get five minutes on stage: one minute to show/explain your script, four minutes to hear people fight about how they would improve it. That would be quality geeky entertainment.

I'm here at OSCon in Portland enjoying the wireless access. I just overheard a reporter talking on his cell phone with his editor (I presume) about how there's nothing "mind shattering" at the conference yet—and stories are hard to come by. He did say he was going to put together a story about the fact that Microsoft is buying lunch for everyone at the conference. You can't buy press like that. Oh wait...

This conference is quite a bit bigger than those I've regularly attended in the past: eTech, SXSW, etc. There are eight sessions or so going on in each time slot, and the trade show has around 25 vendors. I missed the sessions I wanted to see this morning, but I'm hoping to make up for it this afternoon. If you're here I'll probably see you at one or more of the following:

• Intro to Web Services with PHP
• Perl Lightning Talks
• MS Office Files Exposed
• XML Security Standards
• What the X is XQuery?
• Filtering Email with Perl

It should be a mind shattering afternoon. ;)

Highway near Philomath

Barn at the Finley Wildlife Refuge

The weblog landscape is about to change: AOL Blogs. (A preview by Jeff Jarvis.)

I'm looking forward to the Open Source Convention in Portland next week. I primarily use Microsoft tools for development, so this should be a good overview of what's happening in the land of the free. I have been working with Perl quite a bit lately and I hope to catch some advice from the experts. The conference seems to include a discussion of open systems rather than exclusively "open source" software, and open systems are something I'm very interested in. The Emerging Topics track looks great, especially Bill of Rights for Web Services on Thursday.

I'm playing around with Google's new AdSense ads on the snapGallery page. You can follow along by watching the "Ad Stats" box directly underneath the vertical ad banner on the right-hand side of the page. It's updated every two hours with the latest statistics for that banner. Since I added the Google ads on June 24th, they've been clicked 31 times and I've made $3.70. We'll see how it goes.

Of course any profits from this experiment will go toward onfocus infrastructure maintenance/improvement—with the savings passed on to you!

An Amazon developer is keeping a weblog about what it's like to be an Amazon developer. [via Anil] I wonder what sort of weblog policies they have, and how closely this is monitored before/after publishing. What if Amazon gave every employee a weblog on Amazon.com?