Disqus - Latest Comments for dsjoerg

Re: Language of fungi derived from their electrical spiking activity

dsjoerg — Wed, 13 Apr 2022 07:22:15 -0000

Fungi lack chloroplasts and therefore are not plants. These are technical terms, please do not use them indiscriminately.

Re: 48% of Americans delayed care amid pandemic; antibody tests may be wrong half the time — 5 updates

dsjoerg — Wed, 27 May 2020 14:53:05 -0000

No, the CDC did not say "COVID-19 antibody tests are wrong nearly half the time". They said, "For example, in a population where the prevalence is 5%, a test with 90% sensitivity and 95% specificity will yield a positive predictive value of 49%." FOR EXAMPLE.

Re: Pour One Out: Penn Station's Big Board Is No More

dsjoerg — Wed, 25 Jan 2017 15:00:36 -0000

That would mean sometimes they would direct people to track 5 and then, due to trains arriving or leaving differently than expected, need to redirect people to a different track. So then someone would need to loudly alert people that their train was coming in at a different track than expected, which would cause a lot of chaos and upset. Similar to how in an airport, gate changes drive people crazy.

Re: S3006

dsjoerg — Wed, 27 May 2015 20:35:30 -0000

What happened to this bill.

Re: Javascript Chart Libraries

dsjoerg — Thu, 02 Apr 2015 10:15:26 -0000

Been happily using highcharts for years, it's fantastic. Default behaviors are highly intelligent, and powerful customization available under the cover. I have nothing to do with them, just a satisfied user.

Re: The Trouble With Disclosure: It Doesn’t Work

dsjoerg — Wed, 11 Feb 2015 15:38:12 -0000

The article starts out talking about disclosure in general, but then quickly limits itself to the question of disclosure's (lack of) effect on individual transactions such as buying food or entering into a mortgage. Your point, on the other hand, is about the value of transparency to the public's ability to govern, which is quite another matter.

Re: The Wall Street Takeover of Charity

dsjoerg — Wed, 10 Dec 2014 13:52:53 -0000

If only the world of philanthropy were as metrics-obsessed as you suggest, then perhaps the Red Cross wouldn't have had the disaster that ProPublica recently uncovered. Image and anecdote, sadly, rule the day, to the detriment of those who urgently need help.

The swipe against Robin Hood and metric-driven philanthropy is a misplaced grace note in this article that is mostly about donor-advised funds. I hope that you have time to take a closer look at the world of philanthropy in a future piece. I expect that when you look closer you'll find that a) Robin Hood is doing some of the best work in the industry in terms of maximizing the impact of their funds, and that b) far from being a slave to the numbers, they fully recognize the dangers inherent in their use, and treat them as one — albeit indispensable — tool among many in understanding how to best achieve the accomplishments that we all hope for.

An interesting angle for a future piece might be to look at the degree of (non)transparency regarding results in the philanthropic world. Look at the difficulties Charity Navigator faces in building their results-driven report card. In an industry where everyone is a non-profit and everyone has complementary goals, why is there so much hiding and obfuscation?

Re: The Real Roots of Hedge Fund Manager Rage

dsjoerg — Wed, 12 Nov 2014 13:59:07 -0000

In what way was "the Year 2000 bug... the Kim Kardashian of economic problems."? Color me confused.

Re: In Patz Case, a Critical Moment in Court at Last

dsjoerg — Fri, 12 Sep 2014 15:18:03 -0000

Is there any legitimate reason not to record interrogations? Honest question. Obviously there are plenty of illegitimate reasons. I get videotaped on the street corner and at every ATM, but not while being interrogated when my life and liberty are on the line?

Re: Passing vs. Optioning (for VCs)

dsjoerg — Thu, 29 May 2014 16:04:10 -0000

Are you saying that maybe-yes and maybe-no are different? I don't yet see the difference between them functionally — is there merely a difference in emphasis?

Re: OpenSSL Errors and Rails – Certificate Verify Failed – Gem::RemoteFetcher::FetchError

dsjoerg — Mon, 03 Mar 2014 12:49:37 -0000

Thank you! This page saved me hours of stabbing out my eyes.

Re: Privacy Tools: How to Build Better Passwords

dsjoerg — Wed, 22 Jan 2014 08:49:46 -0000

Why do you think passwordmeter is any good? Its criteria seem made up.

The password strength estimator referenced by the article, on the other hand, actually measures entropy, and is dictionary-aware, which is about as good as you can get.

Re: Privacy Tools: How to Build Better Passwords

dsjoerg — Wed, 22 Jan 2014 08:43:30 -0000

To answer your question, no.

I stated my relevant credentials at the end, in a parenthetical, because they are *not* the core of my argument, to which you neither you, nor bigred, have replied at all.

And it's not just "a Harvard degree" — it's a degree in CS. I would happily engage with any argument on the merits, credentials aside. I know some great hackers with no degree at all.

Now go troll somewhere else; ProPublica deserves better than this.

Re: Privacy Tools: How to Build Better Passwords

dsjoerg — Tue, 21 Jan 2014 13:28:51 -0000

KeePass or LastPass (which I use) are a fine way to go.

However, the Diceware method is not nearly as bad as you suggest.

The number of elements alone doesn't matter; it's the total number of possible combinations (aka entropy), which is the product of each element's number of possible combinations.

So with the Diceware approach, there are 7776^4 = 3.66*10^15 combinations possible. That's 3.66 million billion. Even if a hacker has a dictionary and uses the approach you describe, they have to run through those combinations to find your password.

(Since this topic is not simply a matter of one's opinion, I should mention I have a computer science degree from Harvard.)

Re: [Update] Moving Towards iOS 7

dsjoerg — Sun, 08 Dec 2013 19:28:19 -0000

Hi, is there an ETA? This is a deal-breaker for us as we recently added features that break for iOS 6 and before.

Re: Announcing Artillery's Project Atlas, a Hardcore RTS for the Browser - The Artillery Blog

dsjoerg — Sun, 22 Sep 2013 11:21:40 -0000

While I'm proud of my trolling, it's time to try and bring the circle of hate to an end. I shouldn't have been so specific in suggesting the particular remedy of integers. Rather I should have simply pointed out that one doesn't _have_ to use the built-in floats with their associated problems. There are a few libraries that seem to offer solutions, for example: https://github.com/MikeMcl/...

The problem I'm addressing is OP's problem: "non-determinism in javascript".

Re: Announcing Artillery's Project Atlas, a Hardcore RTS for the Browser - The Artillery Blog

dsjoerg — Thu, 19 Sep 2013 17:01:18 -0000

Floats are for noobs, just use Integers within the game model and scale em towards the end for display later.

Re: Announcing Artillery's Project Atlas, a Hardcore RTS for the Browser - The Artillery Blog

dsjoerg — Thu, 19 Sep 2013 14:32:17 -0000

Have you decided on client-server vs client-client interaction model? Can I build my MMORTS on Artillery? Will there be game replays or the equivalent via API?

Re: targa-vs-tod_2

dsjoerg — Mon, 16 Sep 2013 11:37:29 -0000

Love the hallucinations at the end

Re: Gawker is letting readers rewrite headlines and reframe articles

dsjoerg — Fri, 26 Jul 2013 12:25:16 -0000

I am noob but this sounds like Reddit.

Re: Terminal at your fingertips - Emacs Redux

dsjoerg — Fri, 29 Mar 2013 10:04:09 -0000

I was curious how ansi-term differs from what I usually use, which is M-x shell. Turns out there's an article explaining each of them: http://www.masteringemacs.o...

Re: Marcelo Calbucci's Blog: Google is about to learn a tough lesson

dsjoerg — Wed, 20 Mar 2013 07:10:27 -0000

I would say the lesson to be learned is that no service lasts forever, especially not an apparently unprofitable service provided by a firm with many interests.

I don't blame you for using it! I just wish people were less surprised and more accepting when its host decides to shut it down. Actually I believe most people were in fact exactly like that, but a bunch of the malcontents came to this page.

Re: Marcelo Calbucci's Blog: Google is about to learn a tough lesson

dsjoerg — Fri, 15 Mar 2013 17:31:36 -0000

There is a tough lesson to be learned here, and it is being learned by Google Reader users.

Re: Video Of The Week: The Jack Dorsey Interview

dsjoerg — Sun, 10 Mar 2013 09:57:04 -0000

Anyone have a transcript?

Re: http://blog.birchbox.com/post/4963838779

dsjoerg — Thu, 31 Jan 2013 15:33:56 -0000

You made me google "bubble butt silicone pads"