subwindow

I honestly can't tell if you are serious or not.

This chart explains it all: http://finance.google.com/finance?chdnp=1&chdd=1&chd...;

And if you read the article, you would have seen:

"You must have a playful, intermittent form of exercise. And you must exercise. The benefits are profound. Make it fun, intense according to your own fitness and goals, and brief. Brevity and intensity are keys. Intensity means a little burn in the muscle, not heaving and straining. Brevity means you do not release stress hormones."

He agrees with you.

You should only fully trust the opinion of a VC if they are the target audience of what you are building.

Otherwise, they're just some person with a lot of money. The fact that they have a lot of money actually makes them a worse source of advice, compared with your average Joe.

Some people think it is a great PR strategy. That is because sometimes it is. The problem is when you fool yourself into thinking that "sometimes" is "always". Or, at least, "always for me".

Try to avoid Intuit (Quicken/Quickbooks). They have a monopoly, and it is unfair.

I tried for a bit to compete with them, but the banks are complicit and it is impossible to provide proper account import functionality with the way things are now.

That said, try Less Accounting (http://lessaccounting.com/) or Harvest.

I had always thought that Craigslist was poorly designed, but I've changed my mind. This site is, ostensibly, better designed. But I still had to look around for about 15 seconds before I found what I was looking for.

The way that the Craiglist headings are designed is really great, actually. With large headers you lose vertical separation, and this is especially valuable in visualizing the "search | categories | cities" split.

The Craiglist design may not be a happy accident of bad web design. I think it was done very carefully.

37signals has been doing web applications for five years.

They're not some new startup who are building Ruby on Rails applications because it is the hot new technology.

They're building Ruby on Rails applications because they invented it. It is the hot new technology because of what they did, not vice versa.

As far as I understand it, you're talking about dorks, not nerds.

Nerd: Very Smart, socially inept

Geek: Smart, somewhat socially competent

Dork: Not smart, socially inept

That also explains why most people prefer to be called geeks.

The use of deprecated HTML tags and bad practices in general stopped me from exploring Arc further. I found I couldn't hack the HN code without first fixing all of the HTML. Fixing all of the HTML is a herculean task because it's mixed up with all the code, instead of in some kind of template system.

Arc might be a good language for certain things, but it is obvious that PG hadn't made a serious web app in 10 years. Lots of these (framework-agnostic) design decisions have been settled for 5+ years now, and for good reason.

I think the price points should be adjusted a little bit. Very few people are going to want to pay more than $30 for a shirt. This means that for almost all of the shirts, 3/4ths of them will never sell. It kind of looks bad.

Its not as catchy, but $10 + (Shirt Number/4) seems like a better formula- $10-$35.

Probably a little bit, but they're just one site. Google would never put so much influence on one site- it is too gameable.

A single VPS, and it was running completely un-optimized code. It held up fine.

We get more traffic now on a regular Monday morning than TechCrunch ever sent us. Their influence is greatly overstated.

Businesses can be made from niches. Niche business can have social networks. However, niche social networks are not businesses.

You see which way it flows? With a niche business, the social network is a subset of the functionality if it exists at all. If it is just a niche social network, it probably is not a business.

I think his technical infrastructure was ready for it, but his marketing infrastructure was not ready to take full advantage of it.

I've been TC'd once before, and it hit us at the worst possible time- a temporary banner was up for the hours before launch. That banner got more attention in the article than the service. The site held up fine, but >80% of the traffic was wasted. I think this is a pretty common phenomena.

Who says the sharded webheads and databases have to be on different boxes?

As for the linking problem: I had assumed (wrongly, perhaps) that the application in question is one that is friendly to sharding- an application where little or nothing is shared amongst discrete groups of users (companies, teams, etc) and things are sharded at the group level and not the user level. If the application can't handle that, then sharding is probably not the best database scaling option.

When I've planned out sharded infrastructures, the database usually wasn't that big of a concern. The web framework or system architecture are usually the pain point. And with Rails, you have to abuse establish_connection if you're going to have each web head read from multiple shards.

The easiest route I've gone when setting up a sharded infrastructure is to use subdomains and a 1-1 Web:DB setup. Have each subdomain go (either thru a reverse proxy or hardware load balancer) to a different (sharded) webhead. Each web head talks to two databases- the common database and its sharded DB. With this you'll probably want a "common" web head to handle home page traffic and authentication (after they are authenticated they'd get dished off to their shard).

Salesforce.com was my inspiration for this method, and it is probably reasonably common. It probably also has a name, but I do not know what it is.

I had been considering doing something similar in Ghana. I thought about it for a few weeks and kept coming back to teaching. Teaching programming in particular probably wouldn't be very useful, but teaching basic computer skills definitely would be.

If you worked with a small town (provided they had access to reliable power), you could probably have a measurable impact on the local economy. That is really the only way I could sustain myself in such a situation- by being able to see the good I'm doing to a whole community.

WoW was much better at linking together the big rewards than EQ or FFXI were (I've played all three, along with Asheron's Call 1&2, Dark Age of Camelot, Ragnarok Online, Eve Online and Lord of the Rings Online [Good God!]).

With EQ, big rewards were frequently too far in between (especially hell levels and gaps between major "new spell" levels), which led to people quitting when their addiction waned (ie, minor rewards were considered rote, major rewards non-existent or too far away). Also, the end game (at least until Velious) wasn't nearly as developed as WoW's is, and relied on raids held by huge guilds or even the entire server. This led to higher-level players frequently not having access to any near-term rewards, and eventually losing the addiction.

FFXI's problem was with lower levels. The reward for gaining a level didn't seem as important as in either EQ or WoW. Equipment never seemed like a major reward either, as most players essentially rented equipment from the auction houses.

Having spent half of the past 10 years addicted to MMO's, I can tell you:

Addiction via immersion and frequent incremental rewards.

One: you immerse the user in the game environment so that all of their attention is focused on the game world.

Two: using that focus, you provide a system wherein there is the constant availability of incremental rewards. In games like WoW, every single minute spent goes toward achieving a specific incremental reward. You kill a monster: you get a reward (loot + experience), you go to the nearest town: you get new quests. The key is to have a near-term action (< 5 minutes) available to the user at ALL TIMES where they can achieve some kind of small reward. Larger incremental rewards (levels, skills, raids) are available to keep attention once the smaller rewards become rote.

Look at anything that's ever been called addictive- not just video games. I bet that they share these two traits, and their addictiveness is directly proportional to how well they accomplish these goals.

Can these be applied to a facebook app? Maybe. Would you look sleazy trying to do it with a facebook app? Definitely.

Equating education with work is kind of silly. I mean, being silly is obviously the point of the post, but there's no need to be so morose about school. Education is incredibly valuable, and receiving it for free is nothing to sneeze at. In fact, millions of people pay tens of thousands of dollars a year for it after they get done with the free stuff.

Plus, summer vacation is totally awesome.

I don't understand why this post was downvoted. It seems rational and inoffensive. The main post asked for opinions and he gave his- what's wrong with that?

I would go with Rails because there are several options out there to give you a good starting point for a social network.

Try out Insoshi (http://portal.insoshi.com/) and Lovd By Less (http://lovdbyless.com/). Work with each for about a week and see if you like the way the code works and their extensibility.

Also, I'd concentrate on only one factor, at this point- what platform gets you to launch the fastest? Unless you have significant ASP.net experience, Rails will probably get you there faster. Add in the massive amounts of Rails open source code that you can build off of, and I think the choice is clear.

Don't pay attention to people who say "Rails can't scale" but haven't actually built a large Rails app. I run a Rails app that handles >1,000,000 hits per day, and I've spent about 3 hours resolving the scaling issues that I've run into.

All of them. Or, none of them.

Basically, you need to be able to do everything. From graphic/ui design/css, to html, to the MVC stack, to the database, to the system administration and the networking.

For a startup, being a jack of all trades is far, far more valuable than a specialist. How much use will a startup have for JUST a programmer that won't touch the database or the servers? If I'm hiring a developer who is ostensibly going to be coding in Rails all day long, their ability to do design and database work is just as important (if not more-so) than their actual experience in Rails.

Take a look at the stack above, and learn more about whichever one you're least comfortable with. If you've never touched photoshop before, download their 30 day trial and at least familiarize yourself with it (good frequently-used "beginner's" task: making a transparent PNG of something like an arrow). Never set up a linux system? Clean out that old windows box and install Debian (server) and set it up as a file system.

Rinse and repeat.

This stuff is brand new. Obie is essentially blogging the announcement straight from RailsConf. More press will come soon, I'm sure.

It adds an air of authenticity, allowing you (maybe a little bit) to see things as he does. Also, the pictures break up a very long post.

Nope. I keep meaning to attend an SXSW, but can never find the time/money.

I wrote this four months ago in an email to a friend of mine:

"San Francisco is definitely a great area, and you should at least visit some day. There's a vibe there that's completely different from anywhere else I've ever been. It is hard to explain. You just go there and you feel it. Things seem possible out there that would be laughed at here. That's the best way I can put it.

Every city that I've been to has a certain feel to it. New York is rude and busy. Portland is relaxed and thoughtful. San Francisco is ambitious and free-spirited. Atlanta is comfortable and complacent. Atlanta's feel was good for growing up, and it is one that's good for growing old if you're willing to play it conservatively and live a life of moderate wealth and complacency. But if you're not, you should go to a city that has a better feel for what your goals are at that stage in your life.

Does that make sense? I'm by no means trying to convince you to go to San Francisco, but for me it was important that I travelled around and discovered the city that had the right feel for me. Portland was very close, and San Francisco is nearly spot on. I have no doubt you'll be somewhat different. Give it a try though, eh?"

Its like PG read my email and decided to write an essay on it.

Git or SVN using Capistrano.

If you're deploying code, you really should be using Capistrano. Even if it is not a Rails project. It is a world of difference.

Agreed. Any time you type a comment longer than a paragraph or two, you get that message.

This seems like a fundamental shortcoming of the continuation-based request approach. POST URLs should never expire.

There are very few people out there who can offer objective advice based on experience with both frameworks. Most people can (at best) give biased advice based on experience with one of the frameworks.

That said, I've been doing all of my work in RoR and I'm very satisfied. The end-to-end toolset for Rails is pretty hard to beat.