HNHacker News
TopNewBestAskShowJobs

zemaj

735 karma · joined April 23, 2007

Email: hn@zemaj.com
submissionscomments
zemaj··on My experiment with smart drugs (2008)
I also have A,A (Met) at rs4680. I wouldn't necessarily conclude that modafinil would have negligible effect on us based on that study. The study was narrow, it looked at recovery from sleep loss and looking more if dopamine signalling affected sleep recovery.

There's tons of interesting studies about rs4680 (http://www.snpedia.com/index.php/Rs4680). Although I think all the data is still to broad to make any conclusions.

I think I might give modafinil a go just to see what happens.

zemaj··on New Amazon EC2 Instance Type - The Cluster Compute Instance
Not at all. If you start with a dedicated server you can't offload peak traffic as easily. Always going with a virtual server allows you to scale on the fly and cuts costs when not in use.
zemaj··on New Amazon EC2 Instance Type - The Cluster Compute Instance
I don't agree. I've ditched almost all my dedicated servers across all kinds of apps that I run. I now use Rackspace cloud servers as my universal tool for everything I can.

I used to use EC2, but I've found I get significantly better performance (and of course support) at Rackspace for the same price. Also avoid Rackspace cloud sites, performance is pretty awful.

Anyway back to the dedicated server point. If you use virtual servers you can very easily replicate images, spin up copies and scale your computer power on the fly. Case in point, I was running one of my apps on a $10 pm instance on Rackspace. We got a Techcrunch article out of nowhere and the site instantly died under the traffic. I resized the server on the fly and boom, * straight back up. Once the peak died down I scaled the server back down. A couple of months down the track we get near Techcrunch traffic every day and I haven't had to scale the code base at all - I just keep it running on a larger instance. Much, much more cost effective.

* ok not straight. Unfortunately when resizing a Rackspace cloud instance under heavy load it takes ages to run (30+ min isn't unusual compared with 5 min under light load). I've found that the best solution is to spin up a copy on another instance and change the dns to point to the new one. I wish Rackspace would add some way to temporarily block traffic to a server to it can scale. Some kind of "panic", scale me up, button would be great.

zemaj··on Poll: What is your religion?
I think the split is a fairly good indication of how rational someone is. People who consider themselves atheist tend to act more rationally in the bayesian sense.

I'd like to see a split between atheists and agnostics, particularly because the numbers are so high in that category in this community. I would guess that there would be many more agnostics than atheists. I don't think that being an entrepreneur is particularly bayesianly rational :)

zemaj··on Richard Dawkins interviews Steven Pinker
Seems unlikely. While music may be beneficial to a collective, a collective would not be able exist without some form of communication. It appears that the generation and appreciation of music relies on a number of genes that seem unlikely to have evolved purely for music. The selection pressure would almost certainly be stronger on very basic levels of communication before any higher levels of tribal organisation, if purely because they can't exist without the basic selection on communication.
zemaj··on [dead]
Here's another video http://video.allthingsd.com/video/d8-video-facebook-ceo-mark...
zemaj··on Our greatest shortcoming is our inability to understand the exponential function
But it's all considered in a vacuum. Even though the majority of the population don't understand the problem, this is ok. We have an economic system that benefits those who predict impending needs and place themselves in situations to provide for them.

Take a look at the green industry ramping up. In the not too distance future, I think most of us here expect oil replacement to be highly profitable.

I disagree with these doomsday predictions from pure arithmetic. It seems to me that the reverse is true - exponential "problems" provide the core drive for our economy. What drives the entire start-up industry? I'd say it's identifying profitable exponential growth scenarios and providing for them. To me, this explains how primarily capitalist countries tend toward stability, until them become strangled by ever increasing regulation (that's not to say I'm against regulation in general, but I think we need to be as proactive about removing it as we are about adding it [wow, I went off topic there]).

zemaj··on The Twitter Platform
The more I think about this situation, the less I like it.

At first I was happy that the service I work on was not banned by this ToS change. Even though we use twitter data for monetisation, we don't insert data into timelines.

However, when I look at the services that have now been banned, I can't see any warning signs other than that they were competing with Twitter for monetising their data. This is what my service does. Even though it's not currently banned, doesn't it make sense to abandon development now? The best I can hope for it that it isn't wildly successful, so Twitter doesn't consider it competition...

Every time I read Twitter's explanation for the situation, it reads as "we know our monetisation strategy can't compete with third parties in the short term, so we're banning all competition". Hardly conducive to fostering the best solutions, particularly when Twitter will always have the upper hand with their "official" monetisation platform and analytics for resonance, anyway. What's even worse is the the new ToS is still completely ambiguous. Until I saw Peter's post here I had no idea that the ban was only in the publishing end, not insertion.

Of course all this makes sense from Twitter's perspective, but for third parties... that just leaves us on an ever changing playing field with invisible goals. I could have lived with rules and rev share additions, but completely banning competition... not so much.

PS what's the point of this paragraph from the blog post? "We understand that for a few of these companies, the new Terms of Service prohibit activities in which they’ve invested time and money. We will continue to move as quickly as we can to deliver the Annotations capability to the market so that developers everywhere can create innovative new business solutions on the growing Twitter platform." a slap in the face? We understand that we've wasted your time and money, so here's the next thing for you to waste time and money on. No guarantees, no apologies.

zemaj··on Manufactoria - a game about Turing machines
I'm so trying this out on the next person I interview. "You've got 30 mins, see how far you get".

...after 30 mins... "How would you have improved this game?" "Can you make a problem that cannot be solved in this game?"

zemaj··on S4 - $12 for 1 year of 1 TB write-only storage
I just paid for 1TB (really) :)
zemaj··on The Hilbert Hotel
I've never liked these thought experiments with infinity in them. They always have "and so on" in them. But if you're dealing with an infinite set, "and so on" can't complete during the life of the universe (or well, ever). This in itself seems like it solves the problem - you can never get pass the first step. It's an event horizon, literally.
zemaj··on Charles Munger (of Berkshire Hathaway): Elementary Worldly Wisdom
I can't speak for others, but to me the not-so-obvious was simply understanding/exploring the mental models Charles Munger uses to view and manipulate the world.
zemaj··on Scientists Explain Why Computers Crash But We Don't
Downvoted? Wasn't expecting that from hacker news! :)
zemaj··on Scientists Explain Why Computers Crash But We Don't
That headline is terrible.

My take away: If there is an intelligent designer, they'd be an awful programmer.

zemaj··on Ask HN: Pls review my product: Rapid Social Response
I signed in expecting a demo. It looks like it activates the service straight away when using the sign in button. I couldn't see any information or way to deactivate it, so I revoked the oAuth access from twitter.
zemaj··on 6 Million Unfollows Later, Twitter Moves To Silence ManageTwitter
This seems to be a persistent issue with third party developers. If I was bringing out a service to compete for third party developers on my app, I'd add some sort of streamlined approval process. Some way to guarantee that the rug wouldn't been pulled out from under developers 6 months down the track.
zemaj··on The secret to Farmville isn't gameplay or aesthetics
I find the idea of "sociopathic applications" mentioned in the second last paragraph interesting. I wonder what other applications could be included in that. World of Warcraft? I guess even twitter and facebook are in extreme usage. There's no attempt by those companies to stop people excessively using their services, beyond the point where its beneficial to the individual and only beneficial to the company.
zemaj··on AdAdmire - Ads for followers (not tweets)
We've been thinking about ways to monetise our own twitter application (managetwitter.com - over 30,000 users high growth) so we came up with this.

We think it's an interesting strategy - pay to be put in front of people who might want to follow you. It strengthens twitter's network graph & adds value all round.

Feedback welcome!

zemaj··on How Random Matrix theory sheds a light on the laws underlying reality.
In defence of new scientist, their articles are designed to give the average person an overview of the latest scientific research, without expecting pior knowledge in the field. They're not for a technical audience wishing to apply the ideas involved.

I find them really useful for staying up-to-date on a broad range of topics from which I can jump into further research if I find them interesting.

zemaj··on [dead]
Heh, I'll pay that.
zemaj··on Milgram experiment redone: crowd "tortures" man on fake TV show
Cognitive biases don't mean that the brain as a whole is irrational. While we can't always stop the cognitive biases from occurring, we can avoid situations where they occur. For example when choosing a candidate for a job you want to make a decision before comparing with other people's opinions to avoid the bandwagon effect.

However some you can actually mitigate, just by knowing they happen. For example everyone has loss aversion - where we overvalue the loss of something in compared to how much we value gaining from it first place. By knowing that you have loss aversion you can focus on thinking about how much you actually gained when thinking about how much something will cost when you lose it to get a more "real" estimation. For example I'm giving up a business I've been running for years - it's painful to do and something I've been putting off, but I can mitigate the pain but genuinely examining what benefit I've gained, not just making the decision based on the vague "I don't want to give this up" feeling. It works for lots of things; relationships, jobs, etc...

Personally I do it mostly because I like control. I don't like not understanding my decisions. Understanding why I did something makes me "feel" better. That in itself is reason enough :)

zemaj··on Milgram experiment redone: crowd "tortures" man on fake TV show
I don't think they pretend that this is science - it's for a documentary. I also don't believe the point is to "denounce" TV shows. The real point may be clear in this quote from a participant;

"I wanted to stop the whole time, but I just couldn't. I didn't have the will to do it. And that goes against my nature," he said. "I haven't really figured out why I did it."

We think of ourselves as having control over our actions. We're pretty good as justifying most of the things we do within mental models we make of ourselves. However, the reality is that often our actions are driven primarily by external forces. The human brain is full of biases created through evolution that shape our behaviour far more we often like to admit. I think the feeling of disgust that media like this induces is primarily that realisation. We look for reasons why it can't be so.

The good news is that once you start to understand where our biases are we can compensate to a degree, or at least include them in our mental models.

Wikipedia has a nice collection of cognitive biases http://en.wikipedia.org/wiki/List_of_cognitive_biases and http://lesswrong.com/ is an interesting community trying to educate itself on this topic amongst other things.

zemaj··on First hand: Scaling under load spikes with Rackspace Cloud Server
Thought I'd give people a first hand look at dealing with this type of situation. Would really like to know how other people would have handled it.
zemaj··on Do You Follow Too Many People On Twitter? Use ManageTwitter.
I think it becomes most useful if you have an auto-follow back on your account and 200+ following. It starts to become difficult to get much utility from your twitter stream at that point.

I'm interested - what metrics would you want for choosing who to unfollow?

zemaj··on Do You Follow Too Many People On Twitter? Use ManageTwitter.
Yeah, I did (with the help here and there from a few people where I work). Glad you like it. Took a stock image for the background built a simple design and focused on making it usable. Everyone I speak to with 300+ following seems to have an issue managing people they follow. Decided to do something about it.
zemaj··on Do You Follow Too Many People On Twitter? Use ManageTwitter.
It was a tough choice - make the thing run in IE, or spend more time improving the UI for everyone else. In the end I went with the later. I figured most of our audience wouldn't be using IE, and if they were, they probably had Chrome Frame installed. If you're one of the unlucky few who can't run Chrome Frame for whatever reason, sorry!
zemaj··on Do You Follow Too Many People On Twitter? Use ManageTwitter.
1) Spend a week building this app 2) Get an article on TechCrunch 3) Profit! :)

Went through quite a few iterations on the design. Feedback/improvements are certainly welcome.

zemaj··on ManageTwitter - Fast & Easy Unfollowing
Your post happened :)

I did see a decrease in bounces, thanks.

zemaj··on ManageTwitter - Fast & Easy Unfollowing
Interesting. I had one other report of this so added some better error handling. Will have to log this stuff a bit more.
zemaj··on ManageTwitter - Fast & Easy Unfollowing
No, it does not publish anything unless you explicitly say so (double confirm).
← PreviousPage 3 of 5Next →