Comments for Starting Up

Comment on Evidence Rejects Real Name Commenting by Alex

Alex — Mon, 30 Jul 2012 12:44:47 +0000

This is delightfully timely, sam. I’m actually launching a site (agree the people) in the next day or so to completely improve crappy comment threads. Check out the blog. Actual site very very soon.

Comment on I Knew It. by Rahmad

Rahmad — Wed, 06 Jun 2012 07:41:43 +0000

Not really, cops love their huge sarlaies and benefits and most importantly they love their powers and special legal protections against taxpayers. You don’t become a cop to be a hero these days, you do it to have legal protections, big sarlaies and benefits. Their union is one of the unions that is very anti-people. It pretty much controls the city politics because they buy politicians easily with our tax dollars. This system is is just as bad if not worse than then wall street.

Comment on Why was this allowed to be published? [edited] by Nguyen Kok-Suk

Nguyen Kok-Suk — Mon, 14 May 2012 19:24:31 +0000

the best part is the 6 books that he’s written on investing.

oh you mean i can pay you to teach me how to lose money like you? wow!

Comment on Why TechCrunch Should Write Articles About Women [part 1] by Mother Jones Takes On Brogrammers - Starting Up

Mother Jones Takes On Brogrammers - Starting Up — Fri, 27 Apr 2012 14:59:25 +0000

[...] instances of the rampant sexism in the tech industry. I’ve addressed it once on this blog here. I think what’s especially illuminating is that these male programmers frequently use the [...]

Comment on Bad Idea of the Week: Klout by This is Bullshit, and So Can You! « Starting Up

This is Bullshit, and So Can You! « Starting Up — Fri, 28 Oct 2011 15:05:47 +0000

[...] as large as Twitter or Facebook, who cares if you get nasty replies? Contrary to what services like Klout would have you believe, your social media prowess is not dictated by how other people interact with [...]

Comment on [UPDATES] News Roundup of #OWS Park Cleaning Stand-off by [UPDATED] Protesters Being Told To Leave Park « Starting Up

[UPDATED] Protesters Being Told To Leave Park « Starting Up — Fri, 14 Oct 2011 16:26:15 +0000

[...] 10/14 12:25PM: The protesters were not evicted! Here’s a link roundup for [...]

Comment on Bad Idea of the Week: Klout by Hashable Team Blog

Hashable Team Blog — Mon, 10 Oct 2011 14:42:08 +0000

[...] You’ve written that as a QA guy, you’re “pretty much all about finding ways in which things don’t work”. Do you have any tips or tricks that you would you share with aspiring QA engineers on becoming [...]

Comment on Steve Jobs and Occupy Wall Street by Hashable Team Blog

Hashable Team Blog — Mon, 10 Oct 2011 14:42:02 +0000

[...] == "undefined"){ addthis_share = [];}Today, we are profiling a deep-thinking writer who is not afraid to take a stand. A year-round believer in the beard-lovers pasttime, ‘No Shave November‘, our featured [...]

Comment on Bad Idea of the Week: Klout by Sam Gimbel

Sam Gimbel — Thu, 01 Sep 2011 02:52:24 +0000

Hi Aaron, I appreciate your comment–

I’m a QA guy, and we’re pretty much all about finding ways in which things don’t work. Here’s how I’d address your points:
1. I’ve seen plenty of twitter spammers with high K-scores. They aren’t your typical no-follower link spammers, but they do exist.
2. This is only true if they’re a non-frequent twitter user. It’s the only score they can scrape without permission, thus it’s only valid to conclude that a low K-score = low twitter involvement
3. This is one way you can get a high K-score. It’s not the only way, and it’s not immediately clear why a person has a high or low score without actually investigating their behavior. If this is all it did, this would be valid, but it’s not the only criteria it takes into account, and it suffers because of it. However, fewer criteria would make the score “less valuable,” hence the current strategy.
4. “True Reach” is certainly a useful measure, but it’s basically a count, and a bad one at that. It fluctuates pretty erratically and is yet another opaque metric only useful in comparisons where the elements being compared are unclear. Legitimacy is also a pretty subjective term–it requires Klout to focus on a single type of twitter behavior as being “legitimate.” Frankly, there are many “legitimate” ways to use twitter, and many of them are quite casual.

My point is that for Klout to have a good use-case, it would have to be measuring something that can be measured on a continuum. For that, you need a data set that is stable and able to be represented by a single variable. Social media behavior is multidimensional, unstable, with varying rates of change, and thus cannot be represented as such, in my opinion.

Comment on Bad Idea of the Week: Klout by Aaron Biebert (@Biebert)

Aaron Biebert (@Biebert) — Thu, 01 Sep 2011 02:15:17 +0000

Sam, I appreciate you linking to my post in your #4 point (Is the score relevant?). However, I think you make some incorrect points. Specifically this one:

“This last one represents the biggest beef I have with Klout, and is fueled by the fact that there is not a specific use case where using a Klout score makes sense.”

You can’t think of one specific use where Klout scores make sense? I can’t believe that.

What about these uses?

1) Using a lack of Klout score to know someone is a Twitter spammer (Klout screens them)
2) Using a low Klout score (below 30) to know someone isn’t a frequent Social Media user
3) Using a high Klout score to identify people that get retweeted/liked/shared a lot for PR purposes
4) Using the “True Reach” metric to know how active/legitimate a twitter profile’s followers are