thanks

I've known plenty of companies that have produced great software and have gone completely nowhere because they couldn't market themselves.

The 37 Signals crew produces great software but really excels at marketing, yet has never really addressed it in either of their books. Least not more than what one could easily learn from doing a little research on the web.

I'd gladly pay more than the product price total of both of their books if they wrote the definitive book on web marketing.

Putting aside my pet peeves against these guys, I would say the emperical evidence supports my view. On their side, they go themselves, Plenty of Fish and the guy who wrote "The Four Hour Work Week" (which deals with managing a business that is already up and running, so it's not the same anyway). On my side, I've got about 500,000 data points.

When I casually observed Brad getting Feld Technologies off the ground (Brad used to live in Boston), I saw a lot of (ii) and (iii), not just (iii). Plus Brad had an outstanding partner, which makes it a lot easier.

Rails is quite good and very useful. Most of my systems are .Net based, but if they were not, I would go with Ruby and then use the Rails framework. But it is tainted by the 37 Signals approach. Going from the previous version to the current version, they are cutting features. In other words, version X had 1, 2 and 3 features, and version X+1 will eliminate them. Anyone who knows anything about software knows it is really bad to take features away from users, it just pisses them off.

I find the 37 Signals story interesting, since I cannot figure out why so many people have drunk the Koolaid. Why use crap software when there are so many better alternatives?

And some of the things they do are quite good. They are really good at interface design. Their story in "Getting Real" about launching Basecamp without the ability to charge credit cards, since they would not be charging people for 30 days and they could write that part in the meantime, was utterly brilliant. They are obviously very talented guys, I find it a shame they won't set serious about writing great software, they could no doubt make a great contribution. I would love if someone like Google — who has brilliant people and infinite resources — decided to go after them and put them out of business.

These guys are parasites. They refuse to offer decent products themselves but at the same time, due to their being well known, they capture some percentage of the market that should be given to a company that knows how to develop good software.

If 37 Signals published a word processing package in 2010, they would not offer styles, keep with next, footnotes, kerning, mail merge, table of contents, sections, track changes, sequences, bookmarks or cross references, and they would certainly not include Visual Basic for Applications. They would argue that since they do not need those features, and since writing such features is hard and what is most important is that their software developers work a 4 day work week rather than busting ass to develop a great word processor, you do not need those features, you only think you do. They would offer the equivalent of Easy Writer, a program that IBM sold along with the first IBM PC in 1981. IBM chose Easy Writer because it was so underpowered that IBM did not need to worry about canabilizing sales of its more profitable dedicated word processing stations. And like the several dozen competitors of Microsoft Word that tried such a feature-lite approach in the 1980s, 37 Signals would quickly be put out of business.

37 Signals is emperical proof that P.T. Barnum was correct. There is a sucker born every minute.

http://www.jmitchell.me/favorite-quotations

51. When purchasing software, less is rarely if ever more. Rather, more is more. In almost all cases, you are better off purchasing a full-featured, even high software package or service that you will never outgrow, rather than software with training wheels and no features whose capabilities you will quickly exhaust. Switching software packages is difficult and expensive and it makes much more sense to choose sensibly and correctly the first time, even if the learning curve is a bit steeper for the full featured package. Thus, it is in most cases simply dumb to choose packages or services such as 37 Signals, which offer essentially no features and whose limitations you will quickly become frustrated with.

Please write a blog on your own site as opposed to polluting Brad's blog with all of these nonsense comments.