Particle Debris (week ending 7/2)  Juiced, Joost and Goosed [UPDATED]

| Particle Debris

There was a boatload of technical news that interested me this week. I'm finding that a lot of it is coming to me in Twitter because of the people I follow, rather than digging through RSS feeds. That was virtually an overnight change in how I find interesting technical news.

So, there's a lot to cover, even for a short week.

On Monday, Steve Sande wrote up a nice article, "The TUAW how-to guide to iPhone 3GS video recording and editing." I had lunch with Steve recently, and it seems, we are twins, separated at birth. He's also a good guy to follow on Twitter, @stevensande

Speaking of Twitter, the average user has 126 followers, according to The Guardian. The article has some other interesting factoids provided by Twitter's lead engineer Evan Weaver. There's a link in the article to his talk on the architecture of Twitter as well.

Also on Monday, Jason O'Grady published "Five tips for conserving iPhone battery life." a few are obvious, but several others aren't. And he gives the full path in settings to make the easy change. This is good stuff.

Finally, on Monday, I discovered this scary gem from the previous Friday. Google wants to target customers based on FICO credit scores. It's a feature of our age. So long as digital technology and the Internet exists, from now on, people will invent ideas that fly in the face of privacy and good sense. Don't get used to it, figure out new strategies to deal with it.

[UPDATE: 7/13/09.  Google contacted me and pointed out that my interpretation of the article is not quite right.  Here's the Google clarification.]

Google is not targeting individual customers based on their FICO scores.
Rather, the offering outlined in the article you referenced is simply a list
of sites ranked based on the specific demographic they best reach; the
scoring is calculated from aggregate, anonymous data -- culled from opt-in
panel participants -- by Compete. Google and Compete did not attach
individual FICO scores to this data.

The short version of this is that the research has given Google more
insight into the demographic(s) reached by GCN sites, so we better can help
our customers decide which sites are best for their advertising goals. If an
analogy is helpful, the Compete study was a lot like the online version of a
focus group (although an anonymous one)...and the results from that focus
group helped Google to have more information about sites in the Google
Content Network.

 

On Tuesday, David Sobotta, who worked for Apple for 22 years had some interesting insights on his own former uber-boss in light of Mr. Jobs' recent health issues. In my own case, I tend to wonder if Apple can sustain its heritage without its famous CEO, but Mr. Sobotta has other thoughts worth digesting.

Lately, I've been digging into the subject of the Old School Journalism, the demise of newspapers, and the new tech journalism. Here's a scary, lame, example of how Old School Journalists are trying to change the rules to protect their turf. Watch out for more of this nonsense.

I was an early fan of Joost, and it showed tremendous early promise. But Joost as we used to know it is dead. What went wrong was explained by Om Malik on Tuesday. It's a classic Lessons Learned case for any young entrepreneur.

On Wednesday, the BBC had an insightful story about Jonathan Ive. One of the key insights Mr. Ive provided was that, "For a large mulit-billion dollar company we don't actually make many different products. We're so focused, we're very clear about our goals." Mr. Ive also explained why Apple doesn't use focus groups. Don't miss this one.

Also, midweek, Don Reisinger complained at eWeek about the confusing pricing for Windows 7 and compared how Microsoft's ways contrast with Apple and Snow Leopard. But wait! It's not just confusing. It has Vista Ultimate users in an absolute uproar. By Thursday, some Microsoft customers realized they were getting the "Ultimate Shaft.," according to Ina Fried at CNET. Will things never change with that company?

The same day, TV Watch discussed the U.S. Supreme Court ruling on remote DVRs. Basically, cable services would like to put a virtual DVR in their data center and eliminate the service and deployment headaches associated with a physical DVR in your living room. So far, so good. But I suspect they have an additional agenda, and that's taking control away from users who now exercise power with the fast-forward button when commercials pop up. iTunes and Apple TV is looking better and better, folks.

On Thursday, word reached me about an apparent squabble between NVIDIA and Apple. So far, this is just a rumor, but it makes for interesting reading. I know from experience, it doesn't make sense to approach Apple with too much arrogance and hard ball tactics. Apple just goes out and finds another solution, goosing the offender. It may have started when Apple was pissed off about the heat failures in some previous NVIDIA GPUs, the 8400M and 8600M. There's also some legal squabbling going on that could explain Apple's future decisions about product roadmaps. Electronista drew from the source above and added some explanation/speculation.

Finally, could the move by Apple from ExpressCard slots to SD Card slots in the two smaller MacBook Pros have any additional implications, over and above Apple's assessment of customer preferences and usage? The Register hypothesized that and SD card would be more appropriate for a possible tablet and could be a portent. While it's just speculation, there's some excellent introductory material that is a good reference source for the PCMCIA and ExpressCard sizes and performance.

That's it for this week. Wishing everyone a safe, happy Independence Day celebration and respite.

Technical Word of the Week (TWoW)

1. Repuddle (n.) A rebuttal that includes dragging the opponent through the mud. Also a quasi-homonym. Credit: Jeff Gamet

2. Smuggy (adj.) A property of the atmosphere being both muggy and smoggy. Credit: Jeff Gamet

Sign Up for the Newsletter

Join the TMO Express Daily Newsletter to get the latest Mac headlines in your e-mail every weekday.

Comments

Nom

Speaking of Twitter, the average user has 126 followers

Gah!  I wish journos would stop using this construction - it’s statistically nonsensical.

It could mean:

* “The average followers per user is 126”: add the followers of all users and divide by the number of users = 126.  This is useful in a symmetric distribution, but pretty meaningless in a highly skewed distribution (eg lots of people with few followers, few people with lots of followers).

* “The median followers per user is 126”: sort all the users by the number of followers, and the middle user has 126 followers.  This says nothing about the total number of followers, but is the easiest way to get a “typical user” number.

* “The most common followers per user is 126” (ie the mode).  For many large data sets this will track close to the median.  However, some might have multiple “humps”.

My real beef is that “typical X” mostly should quote the median, or perhaps the mode, but generally the mean is misquoted instead.

Log-in to comment