jon_dahl

I think the point is that Git and Hg are extremely close in features/power/etc., not that every SCM is equal.

Not trying to troll here; if you really prefer SVN over Git or Hg that's your preference. But if you haven't given them a chance, you'll do yourself a favor by trying one of them out.

Anyone solve this by setting aside a few half-days each week for manager stuff, and reserving the other time for maker stuff?

Open source to open source, corporation to corporation.

Basically, you can do just about anything with MIT- or BSD-licensed software. But the GPL tries to make everything around it GPL too. If you build software that integrates with a GPL'd library, you have to release your software with a GPL (or similar) license.

So Zed's plan is to offer his software under a GPL license for free (open-source to open-source), and also sell it without the GPL license for pay (corp to corp).

Good place to go for GPL info: the GPL FAQ (http://www.gnu.org/licenses/gpl-faq.html)

The "news" is that the text is now available online, whereas previously it was only available in print.

Details (if you're interested):

This story is 99% hype.

First, there are several major ancient New Testament manuscripts (Sinaiticus, Vaticanus, Majority Text, and tons of fragments). We have fragments from the second and third centuries, and entire versions from the fourth century (including Codex Sinaiticus). This isn't unusual; most of our manuscripts for other ancient documents (Plato, Aristotle, Virgil, etc.) come from many centuries after they were actually written.

Second, these manuscripts often disagree on wording, but they rarely disagree on content. Definitely not "markedly different". They were orally dictated to scribes, not reprinted by a printing press, which is how many of the differences were introduced.

Third, the "no resurrection" business is false. One of the four gospels - Mark - ends with some of Jesus' followers finding an empty tomb, and an angel saying that Jesus had rose from the dead. End of text. It doesn't actually follow Jesus around post-resurrection. But that's different than not including a resurrection, and it's really, really old news.

For what it's worth, I'm at the beginning stages of taking two stalled startups through the Customer Development process. I'll post my success or failure in 6-9 months. :)

So for the time being, I'm quite comfortable with my testing approach. I've given up on "pure" TDD of writing one test at at time. I'll happily write a block of tests and then the code. I'm also happy to write a block of code and then the tests. I can now even cite a study to show that I'm not a complete moron, even if one study is little more than anecdotal information.

The title is somewhat misleading; it's a discussion of test first vs. test last, not testing vs. not testing.

"The board does not set the plans, objectives, goals, or strategy; instead, the CEO develops these, and presents them to the board for discussion, feedback and (possibly) approval."

Typically, the board DOES set the strategy and goals, and management's job (led by the CEO) is to execute the strategy and achieve the goals.

I think it's a myth that freelancing is less stable than a regular job. They both have risks. Your employer can go under, or let you go for political reasons, or restructure you out of work, or get bought by another company. As a freelancer, you have a completely different set of risks, but you're rewarded well to take the risk - such that you typically only have to work 8-9 months/year to make the same amount you'd make as an employee.

The problem with the advice is that it obscures two of the most important qualities that a startup needs: tenacity and adaptability.

First, vocal and influential Ruby developers talk about this. I've heard Chad Fowler, DHH, and Matz all talk about it at major conferences (RailsConf and RubyConf).

Second, I've worked with dozens of clients on Ruby projects, and talked with dozens more, and I've never heard these sorts of issues come up. Outside of the language war world, people don't really care if DHH swears or if an inappropriate presentation is given at a minor conference.

I'd be interested to see other numbers like this if anyone has them.

But an LLC is more flexible and generally simpler to start.

If you're looking to build a small profitable company, I'd go LLC. If you're looking for VC money today, go C-Corp. If you're looking to build a profitable company but want VC money in a few years, consider an S-Corp.

Talk to a lawyer and accountant, of course.

Or does the idempoten-cy apply only to the response from the server, and not the state of the data on the server?

I disagree with you here.

First, your criticism is interesting and well-reasoned.

Second, you aren't making categorical claims (e.g. metaprogramming sucks/rocks).

Third, you're criticizing Ruby from the right position: as someone who has used it in depth, and isn't religiously for or against it.

What I don't like is language trolling, which often comes in the form of claiming that one language is better than another. Unless you're looking at a narrow slice of the language (performance, documentation, expressiveness), you can't just say that one language is better than another.

Unfortunately, even humor doesn't play well on the internet when it is seen as criticism. "5 joking reasons why your technology sucks" brings the trolls out in full force.

Tell HN: "Ruby vs. PHP" is not of interest to good hackers. http://news.ycombinator.com/item?id=554608

Personally, I really wish developers didn't get stereotyped based on their programming language of choice, and communities weren't pigeonholed based on a few loud members.

Each of your examples is a place where Rails isn't appropriate. But each example should be a multi-tier architecture, and Rails can happily cover one or more tier in each case.

For all of its innovation, I'm a little surprised that the Rails community hasn't put together a quality solution for this problem yet. I assume the ORM differences are a part of this - Django's ORM (I think?) defines the schema in the application, while ActiveRecord gets this from the database directly. But does that really make a Django-style admin impossible with ActiveRecord?

* The partners were equal at the beginning, but down the road, their value and contributions weren't equal

* Passive partners were given the same share as active partners

* People were brought on out of loyalty, not because of their value going forward

All three of these problems can lead to instability and conflict.

Partnering is serious business, and it's important to do it right. If people are bringing money to the table, it's relatively simple - value the company and issue shares. But when a company has no value, or if you aren't even sure where the company will go, it's important to vest ownership progressively, based on contribution. Not to say that this is easy, but it's better than just issuing everyone 1,000,000 shares at the concept/seed stage.

Basically, when you pair, you get one better programmer instead of two lesser programmers. Think about the old 10x/100x developer question (http://c2.com/cgi/wiki?DevelopersWithHighProductivityTenxHun...). In reality, it isn't that simple. A 10x programmer won't write simple code (Hello World) 10x faster. On the other side, a mediocre programmer may be unable to do some of the things that a better programmer can do, no matter how much time you give them. So it's silly to compare productivity along a linear scale like this.

Pairing is similar. It's meaningless to say that a pair 2x as productive as 2 developers, or 3x, or 1.5x, because it depends on the task. They might work far faster at a complex problem, but no faster one programmer at an easy one.

That's why my preference is to pair on anything that is moderately challenging (or harder), and to work separately on the easy stuff.