Comcast's DNS servers are terrible. That's why I switched to "OpenDNS":http://opendns.com/ - it's a good thing.

That being said, sometimes OpenDNS is annoying. Sometimes I see their 'typo' pages when I really shouldn't be.

But their recent blog post shows how bad it could be, if you go with "Google and Dell's malware that now comes standard with your Dell":http://blog.opendns.com/2007/05/22/google-turns-the-page/.

I think this is a scaling issue with Google growing quickly. The basic philosophy of Google is good: make money through ads people actually want to see, remember the user is the customer not the advertiser. I like that philosophy, even though it's lofty and idealistic. They should stick to it.

BTW here's an example of Google's "typo squatting pages":http://www.google.com/hws/sony/afe?hl=en&s=http://digg.xom

Comes a new forum fad that's tearing up the interwebs: *rickrolling*.

What is rickrolling? See "this youtube video":http://www.youtube.com/watch?v=7sK3AqFYAWQ to understand more about it.

The rickrolling consists of posting a link to a video of Rick Astley. By following that above link, you would have been 'rickrolled'. Yep that's the whole thing. Typically a poster of the link will pretend that the link is to another video entirely to entice the reader to click on the link and be rickrolled.

Anyways, that is all.

Got a spam today from 'trendio' via Twitter

"http://twitter.com/Trendio_N":http://twitter.com/Trendio_N

Clever trick - add a thousand friends, post some links in your twitter, some percent will click on your links, I know I did.

It's only the beginning.

I've been playing around with Lucovsky's new "Ajax Feeds API":http://www.niallkennedy.com/blog/archives/2007/04/google-feed-api-json.html and I have to say it's a vast improvement from the 'search api', which is really just a widget you can stick on your blog to serve google ads you get no money for.

But the Feeds Api has a lot lower overhead and is cleaner. As Niall Kennedy points out, it's still 5k of bloatware and 2 extraneous requests heavy, but if you're living on the edge you can just point at the unofficial endpoint.

Lucovsky wrote me and warned me about using it:

bq. Be careful about encouraging people to go directly to the /Gfeeds endpoint. The API is obviously very new and this endpoint and its protocol will certainly change, on a whim, as needed. Anyone coding to this raw URL form without proper support from us will likley get themselves into a bit of trouble...

At one point I was using the great 'unofficial' google rss feeds via Google Feed reader. Google feed reader published any rss feed, in a standard format, cached appropriately with auto-added guids. That worked great for months, but then they decided to cut off the service, which was unofficial anyways to be honest, cutting off a bunch of work I had done.

Anyways all in all it looks like Google is going in the right direction at least with their Ajax APIs.

I feel like with two posts in a row I'm being harsh on Google, but I think it might be a sad trend in their "rebrand of Froogle":

http://googleblog.blogspot.com/2007/04/back-to-basics.html "Google Product Search"

At [[Microsoft]], they have gone rebranding mad. Live, MSN, MSN Live Hotmail, Windows Live Messenger: marketing is supposed to tell a story, but not a Chandler mystery.

Naming everything "Google Generic Product" maybe pushes the Google brand, but to me it just makes it more homogeneous and boring.

At [[Yahoo]], they have great projects: [[Flickr]] and [[del.icio.us]] that they have bought, and not gotten behind with their full weight. There's still no great del.icio.us search - i still have to hack it in the URL. I don't even get Flickr photos in Yahoo's image search.

But the "dodgeball guys":http://www.flickr.com/photos/dpstyles/460987802/ leaving Google so publicly, after the dMarc founders leaving, it makes it seem that Google is falling into the same trap: buying products and letting them founder.

Ironically Microsoft has historically handled purchases well, maybe because it's in their roots since they made their fortunes by buying QDOS from Tim Paterson. Google maybe should copy their acquisition strategies and Yahoo's respect for original names rather than the other way around.

I've been thinking lately about how much the fact that [[Google]] owns search does to change the web.

Of course I'm not suggesting that I think Google referred traffic isn't nice, but the fact that it's the only search engine people really use nowadays means that the web is becoming more of a monoculture.

When I think about design of SWiK originally, I put some thought into [[SEO]], but now I put vastly more thought into it, because being listed in Google is a really big deal.

I was reading "this article":http://www.steve-olson.com/when-did-america-become-a-nation-of-frightened-wimps/

the other day, where Steve puts a lot of vitriol in his "Demolition Man" argument that western civilization is on the decline. What struck me as curious is he censored his language in his post, obscuring a letter in swear words with !.

Steve later posts in the comments:

bq. Derek, I use sh!t instead of the real word, because I read that google penalizes sites with certain words.

It's somewhat ironic that someone posting on the value of freedom would submit to Google's grip on the web.

All of this is a function of 1 algorithm ruling the entire web. What if it changes? Entire businesses may fall by the wayside.

Here's a quote from a PBS interview of san francisco porn blogger Violet Blue who was "eradicated from Google":http://www.pbs.org/mediashift/2007/01/digging_deepergoogle_search_sn.html in one of their anti-porn changes

bq. “It’s more than frustrating for small businesses — it’s a death sentence,” she said. “Google has too much power…It’s something that we might not have ever seen in history. It’s not just small businesses but for someone like me as a blogger, I might be naive, but I have this principle of believing in self-publishing on the web. I count on a democratic dissemination of information rather than from the channels we’ve always been getting it from…When Google breaks, suddenly that’s gone and I don’t have that principle to rely on. We won’t have that diversity of voices if we can’t find anyone.”

I don't have any answers on this one. The power law distribution of free choice is going to create a monoculture on its own. This is probably something we're just going to have to live with until people figure out [[del.icio.us|a better way to find things online]]. Until then publishers are going to have two audiences: their audiences and Google's algorithm.

Tim's idea: a code of conduct for blogs so that people who are expressing themselves through blogs and comments on blogs will be 'civil'.

It's a terrible idea, completely against the basic idea of the internet. Let people decide what they want to read, and authors decide what they want to write, and if you don't like what you read, don't go there again and don't link to it. That system works.

Here's a quote that "highlights the problem":http://fakesteve.blogspot.com/2007/04/oreilly-dear-fsj-youre-busted-arsehole.html with the code of conduct Tim is pushing

bq. So think about. Join the movement. For now, though, we're going to have to put your blog on our "black list" and pressure Technorati, Google and others to block your site. We will also be putting pressure on Wired to cancel its sponsorship. Okay? Nothing personal. But we're serious about being fair and keeping the Internet safe. Namaste.

People on the internet are idiots and jerks, anonymity gives people the opportunity to be as lame as they want to be. The good news is that although sticks and stones will break your bones, words will never hurt you. Sure people can push the boundary of that, but if that happens take it offline, take it up with the legal system. There's already a big code of conduct out there: the US legal system. And best of all the first rule in that system is:

bq. Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the government for a redress of grievances.

By contrast the first rule of the oreilly code of conduct is: "we will delete unacceptable comments."

I've noticed that Alexa goes through good and bad periods. I can compare my own traffic graphs to Alexa's and watch a wide variety of sites and look at general trends that wouldn't make sense - like all sites taking a nose dive for a week randomly.

For a month or so Alexa, has been totally wrong. "Valleywag reports":http://valleywag.com/tech/bubble/alexa-error-triggers-crisis-of-confidence-245600.php that Alexa statistics have been off because of a huge rise in Alexa toolbar use in China and India, leading to non-chinese and indian sites appearing to decrease in popularity, whereas in reality they are increasing. Unfortunately Alexa's sample methodology are far from scientific, and so they are vulnerable to external influence to a high degree.

Here's a relevant quote from the article:

bq. The proof is easy. Techcrunch, bravely, publishes its own traffic stats, as measured by a Sitemeter tracker. The site continues to grow, reaching a record 3.4m pageviews in January during the tech trade show season, nearly 60% above its level of August 2006, as the chart shows below. Since November, Alexa's estimate of Techcrunch's audience has declined by 26%; its actual traffic has increased by the same percentage. Alexa is wrong.

Oh yes the buzz has been building and the traffic increasing, "Twitter.com":http://twitter.com is on the rise.

You can already tell that it's popular by its godawful performance, likely a result of having a non scalable architecture for a one-off idea that is really quickly becoming popular, a hallmark of success stories on the interwebs.

"I've joined":http://twitter.com/alexbosworth of course, a little fashionably late but I set my IM and phone to use twitter and I'm looking into the possibilities of its nice little API.

The good thing about Twitter is that it's very simple, which is good news for the web. Also it's mobile, which is buzz friendly. And RSS/JSON? Get out of here, this is rocking.

Twitter is interestingly the exact opposite of their 'real' 'serious' 'web2.0' site that got a bunch of press from TechCrunch and others in the web2.0 peanut gallery: "Odeo.com":http://odeo.com. That site had no real purpose, a product likely of trying to please too many and trying to be too much. With pressure from investors and onlookers and management and audience, it's probably hard to build something like twitter with all of that, it's just too braindead simple, twitter had to be built on the side as a personal throwaway project.

As "Arrington pointedly said when they launched it last July:":http://www.techcrunch.com/2006/07/15/is-twttr-interesting/

bq. But the fact that this is coming from Odeo makes me wonder - what is this company doing to make their core offering compelling? How do their shareholders feel about side projects like Twttr when their primary product line is, besides the excellent design, a total snoozer?

It seems like that strategy worked where 'core offering' Odeo failed, and I'm thinking 2007 will be the year of Twitter.

Peter Norvig has a good "writeup":http://norvig.com/logs-alexa.html today on how [[Alexa]]'s measurement methodology is a classic case of selection bias and bad statistics.

Peter includes some real numbers from various sites and compares them to Alexa's estimates, what's interesting is not that Alexa is off, but that it's widely off even amongst relatively comparable sites.

I can match those sentiments with our own data culled from our logs versus Alexa, Alexa does not track swik.net traffic except in the broadest sense, "compete.com":http://compete.com is more responsible by not doing day by day numbers when they don't have the sample or the methodology to back it up.

Really though even bad information is somewhat better than no information at all, and until Google reveals the secrets of the interwebs we're stuck with a choice between bad and worse.

It doesn't seem likely though Google will be opening up anytime soon, their lauded "Google Trends":http://google.com/trends application they launched as an example of openness hasn't been updated since November 2006 and cries to open it in the forum are falling on deaf ears there.

Recently I've posted about my stories being disappeared from my submitted stories to [[Digg]] list.

I noticed another odd thing today:

"this story has -1 diggs":http://digg.com/programming/Datatypes_of_HTML;_why_HTML_is_great_for_structure

It's 298 days old and it must be hard to do bug testing on the zillions of stories they have in their database, but to me this implies they are doing some weird monkeying with their data or the way diggs are displayed.

Yesterday the [[Wikipedia]] page on Barak Obama was "replaced with this entry":http://en.wikipedia.org/w/index.php?title=Barack_Obama&diff=109344135&oldid=109273026 on the would-be president:

bq. Barak Obama: A crypto-Muslim and militant communist from Illinois.He is currently seeking to destroy the United States by fooling dumb people and minorities into voting him into office as President. His agenda would include trampling on the Bill of Rights, disarming the citizenry, opening our borders even wider, granting amnesty to illegal aliens, and massive income redistribution. A darker skinned, more subtle version of Stalin.

I saw this "referenced on ValleyWag":http://valleywag.com/tech/jimmy-wales/wikipedia-a-clean-welllighted-place-238029.php (name camelcased to annoy them).

Anyways I did a "tracert on the IP":http://www.network-tools.com/default.asp?prog=trace&host=24.178.92.34 responsible for the vandalism, turns out it is decatur-al.gov - a Government address. Some advice to vandals, you are not as anonymous as you might think.

It's kind of a funny thing, other sites would kill for the kind of user contribution that [[Digg]] receives on an hourly basis. People are pretty devoted to the site. One thing I've been noticing recently is that some stories are being smuggled off into the night by Digg's secret police.

One story I submitted the other day was a creative commons account of a switcher from [[Kubuntu]] to [[OSX]]. I, like a lot of other Apple converts, am interested in genuine switcher stories and I thought Digg's Apple friendly crowd would be too. The story was pretty successful even, drawing in a digg every 5 minutes are so, and it soon made the homepage with 35 votes.

But pretty soon my story is gone. Gone from "my submitted stories":http://digg.com/users/alex.bosworth/news/submitted even. It's still in my "Digg Homepage":http://digg.com/users/alex.bosworth/news/homepage stories

Somewhere I imagine my poor story doing hard labor in a Siberian work camp.

Anyways, I posted the "full text of the article":http://swik.net/OSX/Switching+to+OSX+From+Kubuntu to [[SWiK]] (creative commons ftw) and pretty soon there were interesting comments on there.

I don't really mind Digg's longstanding policy of stealth banning stories. If you can't play nicely with the group, go play by yourself, that seems like a good compromise. We use stealth bans on SWiK too to try and remove feedback from spammers. But pulling it out of my submitted stories altogether seems like overreaching. That's my area. Don't touch that, especially don't kill good faith stuff I put in there.

"Chrisek":http://www.digg.com/users/chrisek/news/homepage, a top 10 digger recently went through the same thing, "posting info about a new web2.0 tagging site called 'wis.dm'":http://digg.com/tech_news/Wis_dm_Through_Tagging - now buried and gone from his submissions too.

Eventually we'll add a 'democratic' way to feature interesting projects on SWiK. Normally I'd study Digg as a model for how a great community site can function, but in this case I think they just annoy valuable submitters.

Normally Amazon does really well around Christmas time, for obvious reasons, and then falls back down.

This seasonal deluge of traffic is probably scary for them and tough on their scalability, but it makes for an interesting pattern on Alexa. Unfortunately for Amazon, Alexa shows last year's 2006 holiday season as poor in comparison with 2005, and not too much changed in 5 years of business.

!http://traffic.alexa.com/graph?c=1&f=555555&u=amazon.com&u=&u=&u=&u=&r=max&y=r&z=3&h=300&w=500!

Of course the web itself has expanded a lot I'm sure since 5 years ago, so maybe their growth has increased, and this graph is deceptive as it slopes down naturally.

As a side note, Amazon.com owns Alexa.com. Alexa.com traffic is not available on Alexa.com for some reason, however Amazon.com is.

!http://pbwiki.com/images/logo.gif!

I'm a big fan of "PBWiki.com":http://pbwiki.com - it's a pretty good [[wiki]] but it's real strength is its goal of being as easy as a peanut butter sandwich. That's a huge feature for a wiki, if editing is simple and fast, setting up the wiki should be too.

It has some good grass roots support but not a lot of mainstream hype.

On the other side of the road is the "Wikia":http://wikia.com project that Jimmy Wales of [[Wikipedia]] fame is setting up to cash in on the popular and excellent [[open source]] [[MediaWiki]] project.

(side note: I would put up the funny picture of Jimmy Wales with two girls referencing his alleged soft core porn background, but he has emailed me to ask me to stop using that picture)

Anyways wikia follows a somewhat different path than pbwiki, going for vertical public wiki hosting rather than small organizational wikis. This strikes me as a risky move, being that large successful verticals will be a large risk as they may want to move out of wikia and into their own hosting. Running ads on a large vertical could also be problematic as well.

To get that in numbers, the largest vertical wiki on pbwiki "according to alexa":http://www.alexa.com/data/details/traffic_details?q=&url=http://pbwiki.com/ is peoplemaps.pbwiki.com - but it's a private wiki and only takes 4% of pbwiki visits. The largest vertical wiki on wikia is the Wookiepedia, the main star wars universe wiki. This one site comprises 15% of all wikia's traffic. Only 10 wikis on wikia comprise about 50% of their traffic.

And maybe this leads to more hype, but the same substance: wikia and pbwiki have about equal traffic, ignoring techcrunch spikes.

!http://traffic.alexa.com/graph?f=555555&u=pbwiki.com/&u=wikia.com&u=&u=&u=&r=6m&y=r&z=1&h=300&w=500!

Both of these sites are steadily increasing though, which is good news: wikis are taking over.

Zillow made the headlines in a big way when it was recently announced, and the Wall Street Journal which launched them is still writing about them, (probably thanks to ex-Microsoft ties at the company).

Of course the WSJ is a little more critical recently: "How Good Are Zillow's Estimates?":http://online.wsj.com/public/article/SB117142055516708035-O6WPplch_duU0zq_zhjQaI19vIg_20080214.html?mod=rss_free

And their traffic this year is crazy down:

!http://traffic.alexa.com/graph?f=555555&u=www.zillow.com/&u=&u=&u=&u=&r=1y&y=r&z=1&h=300&w=500!

One thing I'm noticing about a lot of new sites that live or die: poor Google optimization ([[SEO]]). Those Google Maps ironically don't pull in many Google Referrals, and a lot of sites underestimate the importance of Google when it comes to people looking for reference information.

Yahoo Pipes is back up today and I gave a shot at building something in it.

I tried something totally simple: run two [[del.icio.us]] feeds together.

Except it can't really do that.

First of all, Pipes has Google Data as a native data source but not Yahoo property del.icio.us? It's indicative of how difficult it is for services to work together when it's easier to include a competitor's service over your own in house service.

Second, development in pipes consists of dragging around little boxes. If that sounds intuitive, it isn't if you use it if you want to stretch it too far beyond the use cases it's designed for.

An example of some problems with the interface: first drag a source box, like a yahoo search, into the workspace. Then drag a 'for each' box (which despite the promising name really does next to nothing), into the workspace. The for each box will say "drop source box here". Try dragging your source box there. It won't go. You have to know to drag a box from outside of the workspace, ones inside won't work.

Anyways my successful test managed to pipe a del.icio.us feed through yahoo pipes. Except looking at the XML produced, pipes had stripped out what makes del.icio.us cool: all the tags. The authors were also gone. All data except for a few standard elements had been wiped out.

Basically Yahoo has produced a simple filter engine for RSS with a fancy GUI, that destroys a lot of the good stuff in feeds that you want to keep around. It doesn't live up to the hype.

Google Trends is a great tool and very UnGoogley in its openness about search query traffic, despite its severe limitations of having no numbers and restrictions on search query data.

Anyways, it's February 2007 now [[Google]], it's well past time to draw that little line, update your code and show data from this year.

Searches for 'google' on google:

!http://www.google.com/trends/viz?q=google&date=all&geo=all&graph=weekly_img&sa=N!

Helooooooo Google please add 2007.

Oh Wikipedia what is "Huffington Post":http://en.wikipedia.org/wiki/Huffington_Post?

bq. REALITY IS A COMMODITY

j/k

bq. The Huffington Post (often referred to on the Internet as HuffPost or HuffPo) is a political group weblog founded by Arianna Huffington and Kenneth Lerer. Widely viewed as a response to the conservative Drudge Report, it was launched on May 9, 2005 as a news and commentary outlet. Its roster of bloggers includes many people from Huffington's extensive network of prominent "friends". As of August 8, 2006 it was the 5th most popular weblog overall as measured by web links[1] and the most popular "Analysis and Opinion" web site as measured by web hits.[2]

Anyways these traffic stats that are quotes in Wikipedia highlight that Huffpo is not doing as well as it once was. Here is the Alexa graph for Huffington Post this year:

!http://traffic.alexa.com/graph?f=555555&u=www.huffingtonpost.com/&u=&u=&u=&u=&r=1y&y=r&z=1&h=300&w=500!

I have to say I love their web design if you don't scroll down on the page, it's very simple and drudgereport-esque: big words, big images, and a clean white background.

Anyways, this seems to highlight something I've noticed over the past 6 months, that the 'new media' revolution that was hyped by those in said revolution, is not coming to pass, or at least is not coming very quickly.

I would like it to however, I'd love to be able to get all my entertainment via podcast. Except that internet media seems to be following in cable tv's footsteps: lots of crap, some good stuff for niche audiences that aren't served by the mainstream outlets. The key problem for podcast success seems to be a lack of a revenue or recognition model for show makers. iTunes and Google could do a lot here to help this market grow.