Amazon’s SimpleDB question

Amazon’s SimpleDB question: lets say you have a unique ID and a number of indexes on that, perhaps 6. How many rows will fit in one container (= 10 GB)? MySQL’s table limit is derived from the OS and is often 2 Gigs, if I’m not mistaken. So are we saying that a SimpleDB “table” can be then about 5 times larger? That doesn’t seem much of a win… Am I missing something?

Google is finally removing RSS feeds from their search results. Good. They’re leaving in feeds with enclosures (podcast feeds) because it seems a lot of them don’t have an associated html page (podcasting’s history is to blame for this).

LinkedIn doing ok.

A LinkedIn email states: “Fact: 23 people join LinkedIn every minute”, which means 33,120 people a day (pretty good), or about 100,000 new signups a month. LinkedIn is definitely doing some things right, although they’re struggling to give direction to their niche.

Facebook: It ain’t the Web and it ain’t better.

Doc Searls: “Forgive me for being an old fart, but today’s “social networks” look to
me like yesterday’s online services.” Yep, Facebook does indeed have a distinct AOL feel to it. And that’s not something they’ll be able to shake off.

You gotta love the Doc: “it creeps me out when people treat facebook as “The Web, only better”. It ain’t the Web and it ain’t better.”

Youtube gets geolocation wrong

Youtube decides to geolocate me and automatically give me the closest language to my location. Problem is, the closest language they have is German. I am not in Germany. I don’t (really) speak a lot of German. And I *much* prefer my UI in English, thank you very much.

Google gets this often wrong too.  Damn annoying.image

Knol in a nutshell

Google’s new Knol initiative in a nutshell:

  1. Let users copy and paste Wikipedia content (the license allows it).
  2. The content will rank good, perhaps even better than Wikipedia. (Any questions here?)
  3. Make more adsense cash (and give the users an undefined %).

Knol is destined to be a Wikipedia copy-and-paste clone plus a bunch of spam added on top. Plus a huge branding mistake for Google, who used to be the guys who send you to content, not who own content.

Google gives in to the dark side.

When Google bought Blogger, I thought they went too far. Then they bought Youtube, and although I understand why (advertising), it was a step too far again: Google search results are full of Youtube videos. It’s not fair. Now, they’re starting a wikipedia-like site. I am now giving up on Google. They’ve lost their way. They should point to information, not host or own information, because the temptation to promote their own stuff in the search engine is too big, even Google can’t resist. “Dark side” comparisons are appropriate: from the moment Google starts to promote their own content in their search, they’re no longer independent, and the value of their search goes down. They gave in to the dark side, and as any Star Wars fan knows, at first you feel like you’re getting extra power. So you want more. But it’s the first step towards your own doom.

Startups that want to do search and crush Google: you’ve just been given an in. Just say: “We don’t own content and don’t promote our own stuff.” Positioning yourself is suddenly much easier, now that the GOOG has a flaw.

The Bloglines 200 bug.

The “Bloglines 200 bug” (where some feeds start showing 200 unread items all of the sudden even though they don’t have any or just a few new items) hit my new account – I thought I had fixed it by creating a brand new account, but no such luck. I have the feeling the Bloglines people don’t even know about this bug.

I kinda like the simpledollar: “It means that every time we make a purchase that doesn’t have real meaning for us, we’ve added another bar to our prison cell.” That sounds about right.

Intranet SEO

Jorge Serrano Cobos wrote me about an interesting concept: intranet SEO. Making internal stuff show up good in the internal search engine.

I have long had the idea that having departments compete to have their stuff show up on the internal search for certain terms might work. I thought they could perhaps bid on terms. Never tried it though.

Cheap speed.

Google’s advantage is it’s infrastructure. Yes. And it’s used to deliver raw speed. Speed speed speed. Fast matters a lot. Users stay longer, come back more often, and you get more of them. I keep saying that, and my fascinating with scalable architectures kind of comes from that too (although scaling isn’t the same as performance of course). And it has to be cheap speed, of course, coz expensive speed doesn’t help you much, and that’s where Google really seems to have the advantage: cheap speed.

I’m sounding like a drug dealer now.

About dementia and home made bacon.

This stuff is why I still love the internet and blogging after all these years:

“I’m a caregiver for my dad who has dementia, high-blood pressure,
reduced kidney function requiring a relatively low sodium diet, and an
overpowering craving for bacon. So, I started to experiment in making
myown bacon at home using those cryovac’d pork loins. I guess there is
an officially recognized criteria for bacon, so maybe I should call my
stuff something else…but it sure has a bacon look (nice and pink but
not raw, smells and tastes like bacon when it’s nice and crispy, but
has a lot less fat and sodium. So I now fearlessly make him BLTs and
bacon omlettes. He eats, is happy and healthier…and I like that too.”