Slashdot Mirror

Roland_Piquepaille, the submitter of this story, is apparently a publicist. Apparently he is paid to have stories placed in media like Slashdot.

Mr. Piquepaille's affiliations and motivations should be disclosed. Does he pay someone at Slashdot to run his stories?

Also, for me, computing is an important area of continuing interest. Most of Mr. Piquepaille's stories, like this one, have very little real connection to computing, so they waste my time. I'm not the only one annoyed by this, apparently: Here is a script to Hide stories submitted by Roland Piquepaille from Slashdot.org.

Digg.com or Reddit.com are more appropriate media for Mr. Piquepaille. However, I suspect that his stories would seldom be considered interesting enough to be placed on the Digg or Reddit front pages.

--
Most people in the U.S. have no idea how corrupt is the Bush administration. Here's my summary of U.S. gov. corruption. Where's yours?

In the speech, Adam Bosworth predicted that "RSS 2.0 and Atom will be the lingua franca that will be used to consume all data from everywhere" because they "are simple formats that are sloppily extensible."

It's true that many seem to be moving in this direction. For example, A9's OpenSearch is a simple extension to RSS. The Findory API offers simple, RSS-based access to news and blog search results. Yahoo offers a few services through more the more complex Yahoo APIs, but offers many more through Yahoo RSS, including news and web search results.

It seems that most web services may end up standardizing on simple REST protocols using RSS and Atom.

Well, IIRC, Ringo (and other CF systems of the time) was person-to-person collaborative filtering. Amazon is item-to-item. The difference is not just a question of perspective (which dimension you aggregate first) but also one of insight: one the Web, a person-to-person collaborative system (like my newsbot; see sig) has to be "trained" on what individual users like.

Essentially, the user has to somehow give the system enough implicit or explicit "ratings" so that a large enough sample is created and the user can be clustered with other similar users. Sounds awesome, but there's an initialization problem: in order for a user to be "hooked" on the recommendations, they have to train the system. Amazon instead said, forget about users and let's focus on items instead: what items have been "rated" together? cluster those. So, when a new user enters the site and starts browsing, our already pre-computed item relations can tell him what other products to look at. It's a pretty great insight.

Now, I'd think that the initialization problem will hold for item-to-item filtering as well: meaning, what happens when you introduce a new item on the site that doesn't have historical purchasing behavior? that one stumps me, but Greg Linden's (the lead name on this patent) Findory somehow manages that trick for news items which are by definition new all the time (my newsbot went with person-to-person filtering instead for that reason alone).

I don't find any of this obvious at all...

At the time the patent was filed, it was extremely uncommon for systems to make automatic recommendations based solely on the behavior of users. When I did my work at Alexa Internet (which was acquired by Amazon) in the late 90s, I had to solve a number of issues which had not been dealt with, both from an engineering perspective and from a quality of results perspective -- few companies, and no academic researchers that I am aware of -- had both the amount of data and the technical talent required to process it in order to test and refine recommendation systems based on transactional information.

My work in this area became Amazon's "customers who shopped for X also shopped for Y feature." Greg Linden, the first name on this patent, is now doing interesting recommendation work with his site Findory.

--Pat / zippy@cs.brandeis.edu / blog / pics.

Amazon.com - E-commerce pioneer seeking to offer the world's largest selection of products online. for details.

AvantGo - Mobile applications for handheld devices.

DynDNS.org - One of the world's largest providers of free and premium Dynamic DNS services.

Findory - Personalized news and blogs aggregator. Findory learns what kind of content you like by the pages you read.

Live365.com - The world's largest Internet radio website.

Salon.com - Online magazine covering news, politics, technology, art, sex and health; winner of numerous web awards.

Weta Digital - Weta Digital are well known as the special effects people behind the Lord of the Rings films. At his OSCON 2004 keynote, Milton Ngan of Weta Digital thanked some technologies, including Mason, which is used as part of their intranet.

A

AcuTrans.net - Home page for AcuTrans, a company providing an online content management system integrated with transcription services (built with Mason) for business, legal, medical, and self-insured companies.

Adventist.org - The official web site of the Seventh-day Adventist Church.

Alhazred - Progressive music project being produced with open source/free software

Alzabo.org - Home page for Alzabo data modelling tool.

American Lung Association of Washington - Assuring lung health for the people of Washington state through research, education, community service and advocacy.

Apartments - Apartments for rent by RentersInc.com. Free apartment search engine and apartment guide.

arabellan - Web presence of Ryan "Exide Arabellan" Zander, a graphical artist.

astrojax.com - amazing fun and action game - community website with lots of features.

Autismeinfocentrum.nl - Information- and documentationcentre about autism and related subjects in the Netherlands.

AutoSupplyUK.com - Used Japanese import auto store.

B BDO - Austrian tax consultancy

Beotechnic - Company specializing in knowhow transfer

Bikeworld.com - Online retailer, sporting a new 100% Mason-powered site that was developed entirely in-house.

bizjournals.com - Publisher of 41 weekly business newspapers across the US.

BlackSpider - Managed services provider focused exclusively on the provision of e-mail security solutions.

Burma-Shave.org - All of the original Burma-Shave jingles, plus the Burma-Shave Daily mailing list.

C

cibera.de - cibera is an online library site which offers a central access point to interdisciplinary material concerning the Spanish- and Portuguese-speaking area as well as the Caribbean.

Cars - iCarsInc.com Cars for sale. Buy and sell new and used cars online. Your next auto purchase starts right here. Find new, used, classics, sports cars, luxury cars, trucks, SUVï½s and even motorcycles for sale.

Well, this is (half) a shameless plug but my newsbot has been ranking news sources and referring meta-news sources according to "trust" for over 3 yrs now. Findory (no affiliation) does something similar by ranking each individual news story.

Might try using Findory instead. Findory keeps search history for news, blog, and web searches without any account or registration, so you're effectively anonymous when you use Findory. You can also delete your entire history permanently any time you want.

[Disclaimer: I work at Findory]

There's plenty of other people doing search history. My Yahoo Search, My Ask Jeeves, and Findory, to name a few.

[Disclaimer: I work at Findory]

It's not just A9. My Yahoo Search, My Ask Jeeves, Findory, and A9 all have had this feature for a while.

[Disclaimer: I work at Findory]

This is very true. Our startup, Findory, runs on a cluster of commodity hardware running open source software (Fedora, Apache, mysql, perl, etc.).

Our burn rate is very low, improving our chances of survival and reducing our risk.

Open source is a huge help to startups. It's provided a tremendous boost to innovation.

• wouldn't it be cool for the news page to adapt organically in response to stories you click on?

I'm biased, but you really might try Findory. It's a feed reader that learns your interests, searches thousands of feeds, and helps surface interesting articles. It's all about avoiding information overload from RSS feeds.

It's easy to use. Just click on a few articles. That's it.

Well, I am biased, but I think the future of news delivery is towards personalized news delivery, tailored to the interests of each reader. Memigo and Findory (no relation) are two examples of personalized news agents. MSN Newsbot is also going into that direction and I am betting Google will follow soon.

www.findory.com

Exactly. Google's rise to success over AltaVista and others was based on quality over quantity. Precision and recall, traditional measures of quality in information retrieval, don't really matter. What matters is making sure the first few results or even just the first result (the "I'm Feeling Lucky" button on Google) are as relevant as possible. It's all about the relevance rank.

It'll be interesting to see if someone can do the same thing for news. Here's our attempt. Findory learns your interests, searches through thousands of news sources, and helps surface interesting news you'd otherwise miss. It's all about relevance.

Might check out Findory News. Findory doesn't require a login; you're anonymous when you use the site. The design is simple and clean, all text. And the personalization quickly and effectively learns your interests. Try it out and see what you think.

• Who needs 4800 news sources? Too much information = too easy to lose the salient stuff.

Microsoft does excel at imitation. MSN Newsbot appears to be a combination of Google News and personalization technology like Amazon.com or Findory News.

Memigo is very cool with a ton of power user features. It's a fantastic site.

If you're looking for something simple and easy to use, Findory News is another good choice.

MSN Newsbot does look a lot like Google News, but it does have something unusual, personalized news. The site watches which articles you read and attempts to find other interesting articles. Microsoft certainly isn't the first here, but they are the biggest.

If done right, personalized news can work very well. Of course, trust is a big issue. If you don't trust Microsoft, give Findory News or Memigo a try.

• The Times should customize its content so that readers could pick and choose which stories they want based on their own particular interests, rather than having to wade through the site's table of contents.

Well, then check out memigo (my own newsbot) or findory (not affiliated) for interesting experiments on news delivery without Google's infrastructure (both sites are one-man operations, and I only work on memigo on my spare time).

Findory.com (run by a Slashdot user) filter's news based on user preferences. It stores preferences automatically using cookies and require no registration.

If you're curious what a personalized version of Google News might look like, take a look at Findory News. Findory learns from the news you read, searches thousands of sources, and finds articles that match your interests.

In the shameless-self-promotion-department, you might try Findory News instead of MSN Newsbot.

Findory News is a personalized newspaper that learns from the news you read, finds articles that match your interests, and customizes a front page of news stories specifically for you.

• Formerly CRAYON was, IMHO a great site for quick-surfing only the news you wanted to read

• Why not just use slashdot???

• Plus, they HAVE done new things, such as google news.

• [Google] News is my primary source for world news.

Google's AdWords program is remarkable in that truly anyone can buy ads. Small businesses with tiny marketing budgets can buy ads easily. Individuals can buy ads. The interface is simple and easy to use. Google even has a bunch of small business friendly features like limiting your cost per ad and total daily ad spend.

I've bought a bunch of ads on Google, most recently for my startup, Findory News. Most web advertising is expensive, difficult to set up, and performs poorly. But, because you can pick such specific keywords with Google Adwords and the advertising engine refuses to show ads that don't perform well, you can easily get in front of people that might be interested in them.

It's offtopic, but I'm curious what suggestions you might have for a "face lift" for the site, especially since you're a frequent user of Memigo. If you'd be willing, please get in touch at the e-mail address provided on the About Findory News page. I'd very much appreciate hearing your comments. Thanks, cpeterso.

It's a little offtopic, but I'll briefly respond.

Findory News isn't the same as Google News. Google News does pull from a broad variety of sources, but it shows the same news to everyone. Findory News is personalized. It learns your interests and emphasizes the most important news for you. More information is available in the about and FAQ pages.

It's a little offtopic, but I'll briefly respond.

Findory News isn't the same as Google News. Google News does pull from a broad variety of sources, but it shows the same news to everyone. Findory News is personalized. It learns your interests and emphasizes the most important news for you. More information is available in the about and FAQ pages.

It's a little offtopic, but I'll briefly respond.

Findory News isn't the same as Google News. Google News does pull from a broad variety of sources, but it shows the same news to everyone. Findory News is personalized. It learns your interests and emphasizes the most important news for you. More information is available in the about and FAQ pages.

We're getting a bit offtopic, but I'll briefly respond to this.

Advertising revenue cratered in 2001-2002, but it's been picking up again. Google, for example, is estimated to have nearly $1B in ad revenue in 2003. Google's technique, in Google AdWords, is to make advertising targeted and useful. They do text-only ads, but they achieve much higher clickthrough rates than obnoxious popups or take-over-the-screen ads that you see elsewhere because they're actually relevant and helpful to Google's users.

I'm not going to go into how this technique could be applied to a personalized news site like Findory News, but I'm sure you can imagine some possibilities.

This looks like it though.

I'm not sure it's that easy.

The biggest problem for unemployed software engineers getting together and forming a company is financing. If you're unemployed, you probably don't have a lot of cash around to provide seed capital for your business. If you do scrape together enough seed capital or find an angel investor, you're usually stuck with not drawing a salary and having no health insurance for about a year as you try to ramp up your business. And you're doing all of this for a high risk venture where only 1 in 10 businesses actually succeed.

Nevertheless, I'm building a startup right now, Findory.com. But I wouldn't underestimate the obstacles here. It isn't something that can be done lightly.

It's a bit of a pain to get news from a wide variety of sources though. Takes quite a bit of time. I've found a couple resources to be useful for this.

Google News and Technorati pick from a wide variety of sources and allow you to search news articles. But I sometimes find it hard to find what I want in all the clutter. Plastic provides moderation and discussion of news, but doesn't have broad coverage. Various RSS aggregators allow me to create your own news feed, but they don't have good coverage of mainstream news sources and they're a bit of a pain to set up. There's a couple recent attempts at personalized news -- Findory News is one -- that try to pull news from a broad variety of sources targetted to your interests.

What do you use?

It seems to me that search could be improved by paying attention to what you've done in the past. Current search engines assume each search is completely independent, but there's valuable information in what you did earlier, especially if you keep refining the query trying to find something. One attempt at this is demoed at Findory Search. It particularly makes a difference if you're refining your search query to try to find something. But it's still a work in progress.

Memigo is very cool. Findory News is worth seeing too. Both are interesting examples of a personalized news site.