zealog

While it is interesting that there is "no overtaking" in the ant world, I don't think this really has any real bearing on humans and driving congestion (as suggested by the articles subheading). Having sat in long, single lane stop and go traffic I can anecdotally suggest it's irrelevant.

More importantly, there is a big difference between personal locomotion and commanding a vehicle. For example, the difference between my walking pace and that of an Olympic champion sprinter is not THAT great. However, the difference between my freeway merge speed and that of my mom is an order of magnitude. Add in fixed and inflexible travel lanes, the ability to speed up and slow down consciously and unconsciously, traveling at speeds that will knowingly cause death in the event of a problem, etc. and you have a lot more variance with higher stakes to the participants.

I'm not saying it isn't a fascinating bit of trivia, but I think Tom Vanderbilt's book "Traffic" is apt to lead to more insight than this particular nugget of knowledge.

Like if you're as rich as Bernie Madoff you must be legit?

You do realize that the "con" in "con man" comes from the word confidence, right? The most important thing for any scam to work is building confidence.

What better way than with celebrity-ish pitch men and a Superbowl ad?

I agree.

I've been using NGinx/Mongrel for most of my large deployed apps, but am working to get things moved to Passenger (and Apache, natch), simply for the ability to do a graceful restart on most code changes. For large apps, the ever expanding mongrel footprint and slow restarts is becoming too much to bear.

So for me, Nginx has a lot to offer over Apache, but with the rails deployments I've been working on mostly, it's no longer enough.

Though, for me, it results in fewer visits to the site.

I would venture to guess that each of those clicks are less valuable than 25% of a single click, but many publishers are always looking to fleece their advertisers any way they can.

I sincerely doubt that Google fears Twitter, but twitter search can and does provide something very powerful that, at this point, Google does not. That "something" is real time info.

Some examples:

1. Some Saturday Night around midnight try searching Twitter for SNL. (You don't really have to search, it's usually a trending topic). You will instantly find out who the guest is, if there is anything particularly funny, if the monologue bombed, etc. Go ahead and say "SNL is never funny," but it is just an example of real time feedback. You can do the same thing during awards shows or sporting events.

2. Since silicon valley has a relatively high density of Twitter users, I will sometimes find myself doing a twitter search for the road I'm about to take if I'm concerned about traffic. Sure, I can see traffic on Google maps on my iPhone but it only covers freeways and is not as fast or granular as people complaining that they aren't moving for whatever reason at a paricular intersection.

3. Speaking of iPhones, Twitter search was AWESOME for figuring out where lines were long or short. On the days following the 3G launch, doing a search for 'iPhone line' would produce results like "About 200 people in iPhone line at Valley Fair - 10 minutes ago" or "iPhone line was only about 30 minutes in Palo Alto - 1 hour ago." Quickly getting a sense of the crowds before driving all over town is VERY useful.

Google, while definitely not threatened, has no way to provide me that kind of real time info, and until they do, Twitter has an important place in search.

For me, the money wouldn't do it. I need competition. It actually works and I was winning until yesterday. http://zealog.com/zealogs/the-smaller-losers

They are animated gifs (like ajaxjoad.info), but there are some nice and original spinners there.

The smartest thing about this post is the point he makes about developers re-evaluating their tool chain at the end of each project. Anyone who makes tools needs to keep that in mind.

I completely do that - and it's how I ended up using (and loving) jQuery. Of course, it makes going back to work on old projects a drag and inspires a lot of (too much?) refactoring.

Very nice. I will definitely be using it in the future. Very near future, I'm sure.

I'm not sure they've failed, but their lack of traction has to do with 3 things, I think.

1. Trying to do too many things. Too many features meands market confusion and cluttered interfaces. There is power in simplicity and purpose.

2. Twitter had already won. The early movers and adopters were already there and the network effect was in motion.

3. Using Adobe Air for their non-web interface. Twitter never worked for me until Twitterific, but air is just SLOW and annoying I personally couldn't abide.

This looks great, but I'd be hard pressed to use it on anything other than occasional headers until someone figures out a hack for copy/paste support.

A pretty good article, but I heartily disagree about not using a conditional style sheet for IE.

I develop for FF and Safari on a Mac. When things are how they need to be I run through the various IE versions and make sure they work like a real browser as much as possible.

I much prefer knowing that all of my IE hacks will ONLY be shown to IE browsers and I don't run the risk of screwing up the layout for non-IE browsers because some hack has leaked through. Besides, it is much more readable to have the styles as intended be as simple as possible as opposed to ,things like..

  div.middle {
    min-width: 300px;
    width: auto !important;
    *width: 300px;
  }

There is nothing dumber than wading through stuff like this when you aren't specifically building for IE.

I tend to go either way when building a site, depending on the market and how people are going to interact with the site and with others. On sites with kids, using real names is a BIG turn off and if it isn't clear how the name will be used, people will pass.

It's anecdotal, but I use my real name in many places (not ycombinator, ironically), but my wife never will. If she can't choose some sort of display name, she will pass or use fake info.

Also, you might want to check out Smashing Magazine's recent survey of signup web forms. They touch on this a little bit. They don't give much other than statistics, but it might lead to some good discussion. http://www.smashingmagazine.com/2008/07/04/web-form-design-p...

I think this is an excellent point and really gets into what David was talking about in his (absolutely inspiring) Startup School presentation and how it relates to this post.

To many people, being in a startup is about selling the same thing to the same people - eyeballs to VCs or Google. That's it. There frequently isn't a lot of concern over what exactly it is that draws them there. In these situations, long hours are required (or at least mandated) because the target moves so frequently because someone has a "better idea" or needs to implement "just this one thing" to appease some VC who ends up not funding you anyway. This lack of focus and constraints leads to middle of the road products that can only be worked on by people working 14 hour days to adapt to the ongoing changes.

It's sort of like when Hollywood is out to make a "blockbuster." They buy a popular comic book character, get a dozen hack writers to work on it - even while the scenes are being shot - and they put in a bunch of crap special effects. Sure, sometimes it works, but most of the time it's a bunch of dreck that will never make back its $200 million budget. Contrast that to the lifestyle business - er, independent film - being written and directed by the guy who LOVES what he's doing and making it for himself and you get Pulp Fiction.

Me, I'd much rather make "Pulp Fiction" than "Daredevil."

From experience, a private beta signup page leads to a pretty decent bounce rate.

The title is hyperbole, but the core idea - recommended search by contacts - is what EVERYONE is trying to do. Even Google has admitted to such.

I don't know that Twitter will ever be able to make use of it, but there are some bones there (mentioned in this article) that represent the next great frontier in better searching.

That looks absolutely great. Going to have to download now.

For developing apps that run on a Mac, XCode.

For web apps being written on a Mac, TextMate.

I knew that I could do something like [:admin, @record] and the path would know where to go, but I didn't know I could do something like an edit path with edit_polymorphic_path[:admin,@record]. Very handy indeed.

I've read about this bit of the book in other articles. The cheating was not on the books/10 commandments, it was on a math quiz (that afforded cheating) which followed the listing.

For those that listed books, some would cheat on the next bit. For those that listed commandments, they wouldn't.

The article is very interesting on the psychology of how it works and anecdotally I would completely agree with the findings.

However, the title is putting the cart before the horse. Just because people respond to things this way, it is not the cause of the tradition.

The best researched article I could find (that matches a story from an old book I once had) about the origins of the $.99 can be found here http://www.straightdope.com/classics/a3_166.html

Nice overview for a first timer, but please tell me the designer is sending PNGs and not JPGs of a mockup.

http://twitter.com/zealog http://twitter.com/aaronh

Cool. Simple. Clean. Thumbs up.

Great article and lots of good advice.

The one about cheerleaders is important. I'd hemmed and hawed around my idea for months, but once I started building it, using it myself, and had a few friends as alpha testers, it became much more exciting to work on because I was making it for me. I could see the incremental things needing to be done to improve it as opposed to just a giant mount of unformed and uncompleted toil.

I also like the idea of using wasted time to at least learn something. Of course, the danger there I face is reading a new blog post that more elegantly solves a problem than I recently had and wanting to go back and refactor (or worse, rewrite from scratch). As long as I can keep that in check and moving forward, it's great advice.

"The thought of wasting 4-5 months of my life like that scares me."

I get that. It is incredibly difficult when you look at it like that. I have personally felt that pain of, "So, who I do I take time from, my wife or my kids?"

However, in the case you mention, it is only 4 or 5 months of nights and weekends. What is that, 30 actual days max? Would you trade 30 days of the thousands and thousands in your life to be able to make something you are proud of and potentially allow you to spend all of your time doing what you love? I know I would.

Obviously, thinking like that can lead to a workaholic, Cats in the Cradle (google it, young'ns) type existence, but with measured use of it and diligence at maintaining boundaries, I think it is possible.

I remember when I had a friend ask to me run in a 5K. Before that, I wouldn't run unless being chased, and even then... However, I realized as much as a pain as it might be, even being slow 5K is about 30 minutes. No matter how painful, annoying, embarrassing, it is, that's half an hour. 2% of 1 day. I decided right then that I could do anything for 30 minutes - and did. This is the same sort of thing.

You can do anything for 4 or 5 months and see what happens. This IS your life, and it's ending one minute at a time.

</pollyanna boosterism>

I saw this the other day and thought it was pretty neat.

I'm a little unclear how it works. Does it actually attempt to load a page from the requested site or does it simply log the number of people asking about it and assume that something is down by the volume of requests?

If the 2nd one, I can see how it could be pretty flawed, easily gamed, and only usable with some serious critical mass. (It could also explain the YouTube thing if lots of people are using that to check it out).

For me, it's all about number 2 (and how it relates to #1) and my current contracts and obligations at my "day job."

I build startups all day every day, so I'm comfortable with the field. I've got what I think is a good idea and a ton of execution on it so far. I have a small group of testers that (about 20% friends, the rest random strangers) and a desire to make a great product.

There is a great monetary upside potential, but I just don't have the resources at this time to quit and focus 100% on this. Of course, I sleep so very little that I spend at least as many hours "on the side" as I do at my day job, so it might be somewhat moot anyway.

I really think this is the key point and problem. If the people doing the heavy lifting aren't either passionate about the product or actually using it on a daily basis, you will suffer.

I don't think you're doomed, but as a programmer, I've both had startups and been brought into ones that have been outsourced. In my experience, those little things, the fit and finish, just aren't there when the people building them are in it for the paycheck.

At the risk of sounding cliched, it's kind of like the old adage about people who are fighting for their home vs. a paid soldier. You will get a lot more fight from the guy defending his home than the mercenary.

I can't directly answer your question I have only used shared hosting, dedicated servers, and EC2.

However, I would say that EC2 is actually not that expensive compared to the others. I've been using it for my latest project and really like it. The cost is only about $75 a month and it behaves just like my dedicated servers (but it has more RAM!). That's coffee money.

Also, you will eventually leave the "development stage" and finding a place where you can easily (and without great expense) move to a larger VH will save you money in the long run, too.