Disqus - Friends of AnthonyPhan

Re: Happy birthday Foucault’s Pendulum

Eric — Thu, 08 Jan 2009 19:27:14 -0000

Then you have people like Barry Marshall and Morris on Helicobacter pylori. Marshall drank a culture to induce ulcers (and thus show a causative relationship between the bacteria and ulcers), and then Morris, in the name of scientific reproducibility, did the same thing to himself and published the confirmation. Marshall recovered with antibiotics, but Morris remained debilitated for about 3 years...

Re: Happy birthday Foucault’s Pendulum

Eric — Thu, 08 Jan 2009 20:39:03 -0000

Ah yes, Marshall's experiment probably wasn't so elegant. Intense pain, GI problems, etc....not so clean...

Re: The Promise of Imaging Mass Spectrometry

Eric — Thu, 15 Jan 2009 16:51:23 -0000

Anthony's right. Mostly it just does x-y positioning. There is a way to get z-positioning by looking at how the molecular profile changes over time (i.e. as the laser blasts down deeper into the sample), but that's not so accurate, it doesn't work as well on non-homogeneous samples (because the rate at which the laser blasts through the sample changes from place to place), and because you can only get so deeply into the sample. Besides, it's just easier to make slices of the tissue and reconstruct the model in 3D afterwards.

Re: Joe

Eric — Tue, 17 Mar 2009 07:47:37 -0000

I think it would be nice to have a snappier and more responsive interface without the slow fade ins and outs. Also, having the dotted lines disappear when the mouse is over the picture means that in order to see what the other targets are, you have to move the mouse out in order to move the mouse to a different point!

Re: No Money for Blockbuster

Eric — Mon, 30 Mar 2009 14:50:33 -0000

Not only are Blockbuster's retail locations irrelevant because of the internet, but they're also not quite as convenient as the RedBox DVD rental kiosks that are placed in many major supermarket chains.

Re: There is no security

Eric — Mon, 30 Mar 2009 14:52:20 -0000

How do they infiltrate a Blackberry that's turned off? Does that have anything to do with exploiting Blackberry's remote corporate IT services, e.g. remote wipe?

Re: Stuck between a big company and a startup place

Eric — Thu, 21 May 2009 13:00:05 -0000

I think there's a perfectly good reason startups favor execution and boldness at the expense of strategy and analysis: risk. Most of them fail, and most probably fail for lack of proper execution. Since things can die so easily in just the planning stages, better to follow through and then see if things work rather than spend a lot of time (in which you're not getting paid) forecasting.

Re: Innovator’s Delight

Eric — Tue, 22 Sep 2009 10:46:45 -0000

Interesting review, although his estimate for a fruit fly's lifespan is off by a few weeks.

Re: What is with Microsoft's consumer electronics strategy?

Eric — Thu, 10 Dec 2009 14:44:00 -0000

The Windows Mobile vs. Zune thing is particularly egregious, especially considering that Apple is showing them the way. The iPod Touch is a huge driver of present and future iPhone purchases, as well as development of the "ecosystem," especially since it's cheaper and is growing faster in per-unit sales than the iPhone.

I do, however, see how the Xbox fits into the Microsoft strategy, since on the development end it drives Windows PC gaming as well as Xbox gaming. In addition, the Xbox-as-media-center has a common marketplace for content between the Zune, PC, and the Xbox, at least for non-application content.

Re: What it will take to get me to switch to Chrome

Eric — Fri, 08 Jan 2010 12:28:58 -0000

With regards to Firebug (I haven't tried it, so I don't have a feature-by-feature comparison), you do realize that Chrome has a Web Inspector that has many of the capabilities of Firebug (the inspector comes with the WebKit framework, so it's already in Safari as well). Just right-click on any element while in Chrome and choose "Inspect Element". You can profile, edit, debug, and execute JavaScript on the fly, browse and edit the DOM/HTML and CSS, browse and profile page resources and HTTP requests, and also examine HTML5 local databases.

Re: What it will take to get me to switch to Chrome

Eric — Fri, 08 Jan 2010 23:24:02 -0000

I find both interfaces equally clunky in their own ways. I do agree with Joe about the quirks in Web Inspector, but in terms of overall UI quality and polish, and for browsing elements, Web Inspector blows Firebug out of the water.

For example, at least for me, if I highlight an element in Firebug and scroll with two-finger scrolling or my mouse scroll wheel, the outline box doesn't shift with the page correctly. If the sidebar containing style properties in the HTML view is too narrow, the text starts overlapping. Firebug's "Net" panel's time measurements don't have tick marks for time as Web Inspector does, making each timebar's subdivisions a little less useful. Web Inspector lets you re-root the HTML code blocks, so that instead of having everything in a deep hierarchy be scrunched up against the right side, I can just double-click a node and have it show me just that subtree. Also, I like the way the Web Inspector's highlighting looks better. :-P

On the other hand, Firebug has, as Joe said, the Inspect Element feature (Web Inspector's seems buggy) and a better CSS and HTML editor. Web Inspector makes it a little tricky right now for CSS, not because of the separate property/value fields (which I actually don't mind), but because it's a little screwy to add new properties to existing elements. You have to add a semi-colon manually and enter a new one, which is not so elegant a solution. Also, you can't tab forward to edit other things in Web Inspector as you can in Firebug; you have to doubleclick on the thing you want to edit, which is stupid.

I think it depends on what you need. Web Inspector is better at browsing the elements and getting a broad sense of things, while Firebug is better for editing your page in the browser itself.

Plus, I just despise Firefox, which has an ugly UI and is the slowest browser on the Mac by far. :-P Chrome is nice, but I currently don't see much of an advantage over Safari.

Re: Crazy Y Chromosomes

Eric — Mon, 01 Feb 2010 12:36:58 -0000

I think a lot of the claimed similarities between human and chimp genomes doesn't take into account structural and copy number variation, especially since we're discovering more and more intraspecies variation as we analyze new genomes for more than just SNPs.

For example, human chromosome 2 is a fusion of two chimpanzee chromosomes.

Re: Net Neutrality 2.0

Eric — Mon, 12 Apr 2010 19:39:43 -0000

That 3D chart makes me want to claw my eyes out. Most egregious use of 3D I have ever seen.

Re: My brother the master coder

Eric — Wed, 28 Jul 2010 12:20:42 -0000

I have mixed feelings about exceptions. They interrupt the programming flow in the same way as goto statements, and it makes it hard to predict what will happen in the program.

Re: My brother the master coder

Eric — Wed, 28 Jul 2010 13:01:12 -0000

I think the flow is interrupted more if you have the blocks as large
as possible. I'm talking about following logical/execution flow. If
you have more blocks, at least you stay local. Also, exceptions (at
least I've heard) can seriously hurt performance in that you jettison
the CPU pipeline; this is less of a problem in the parallel processing
present, but probably still important.

Some APIs pass back error variables (by reference or by multiple
return statements). Perhaps we should, too.

-- Eric

Re: Why I Favor Google over Apple

Eric — Mon, 23 Aug 2010 09:50:28 -0000

I'd say your broad generalizations are a bit *too* broad. After all, Apple has built all of its operating systems fundamentally on open source components to which it has contributed (e.g. FreeBSD, Mach kernal, Darwin, LLVM, clang, gcc), as well as enabled strong support for many widely used community-developed languages, such as Ruby, Python, Perl, MySQL, etc. I'd say that Apple's support of open-source isn't really an exception to the rule as much as just not being a priority in and of itself. I'm sure ceteris paribus Apple and its employees would prefer to support open source development (free labor? wider adoption of their technologies? who wouldn't?), but they'll only expend company resources to do so if it really means helping their overall strategy of making money.

Re: Droid 2 vs iPhone

Eric — Mon, 30 Aug 2010 11:34:56 -0000

Just FYI, the iPhone can control the music player and phone via Voice Control.

Re: Apple TV Disassembled

Eric — Mon, 01 Nov 2010 09:35:36 -0000

"Who said I wasn’t a comic book nerd?"

No one. No one has ever said that.

Re: Degradation Situation

Eric — Fri, 19 Nov 2010 17:58:30 -0000

"Because then small changes in leak rate or the water flow won’t have such a big impact"

That's not really a good analogy, because high leak rate from a bathtub doesn't translate to the leak increasing proportionally when you add a bit more water. They're not measuring overall degradation, but rather the rate constant, which is more like measuring the steepness of a hill, rather than the height of the hill. A more abstract, but more accurate analogy would be that a ball sitting at the bottom of a narrow, steep dip in the ground would be more stable than one sitting at the bottom of a shallow, rounded dip in the ground, even if the two dips end up being the same depth.

Re: Degradation Situation

Eric — Sun, 21 Nov 2010 21:32:20 -0000

Yes, but faster water flow overall doesn't necessarily mean that a particular extra amount of water results in faster water outflow.

Re: Enough Is Enough

Eric — Wed, 01 Jun 2011 08:58:45 -0000

You either didn't read his comment, his post, or are just not comprehending the distinctions he made. He's referring to intellectual property rights for software, not property rights in general.

Re: Enough Is Enough

Eric — Wed, 01 Jun 2011 09:01:54 -0000

That in biotech/physical sciences, one invents the design of a concrete physical object, rather than a process or an idea. It is much easier to tell (for a patent lawyer) whether the structure of a drug violates one patent or another compared to telling whether one violates a software or business process patent, and thus it doesn't necessarily cause a chilling effect on a broad swath of innovation.

Re: Enough Is Enough

Eric — Wed, 01 Jun 2011 09:44:51 -0000

From his original comment: "i don't believe someone who comes up with the idea of a transaction in an app has property rights to that idea. and i do not believe that idea is central to capitalism"

Where in there is a general statement about property rights per se? Let us parse: the antecedent of "that idea" (before "is central to capitalism") is most probably "someone who comes up with the idea of a transaction in an app has property rights to that idea." Notice that he has qualified his statement about property rights, so that the antecedent most likely isn't "property rights" in general, but rather the entirety of the causal chain that someone formulating the concept of in-app transactions thus can obtain property rights for that particular concept. Presumably, he being a VC, he believes very strongly in other forms of property rights, owning equity in many different companies.

Therefore, using grammar, a not-so-ridiculous supposition of his internal logical consistency as a venture capitalist, and looking at the overall context of his other comments and posts, you can see that he doesn't say property rights aren't "core to capitalism."

Re: Enough Is Enough

Eric — Wed, 01 Jun 2011 09:58:41 -0000

Meh, I just find blatant misreading of what is actually a plain and clear statement to be annoying. I'm a scientist, my wife's a JD, it's par for the course.

Addendum: I only felt a compelling need to spell it out explicitly because 1) you seem like you're experienced enough to know what you're doing, and 2) it seemed strange that you were misinterpreting Fred's comments so wrongly.

Also, I have this syndrome. http://xkcd.com/386/

Re: Enough Is Enough

Eric — Wed, 01 Jun 2011 13:36:59 -0000

Inevitably, any idea has a physical component as well as an information component, but if the program's hardware implementation is general (as most useful programs are), one could argue that the hardware implementation is trivial and therefore not novel. The distinction isn't an easy one, like most things in law, but one needs to be able to draw a line somewhere for patentability between a math algorithm (not patentable) and a new molecule (patentable).