Hacker News new | comments | show | ask | jobs | submit login

I think Jason Fried and DHH might take issue with, "You proved yourself wrong, I think."

In addition to "Say no by default," one of their other points of advice has been: "Let your users outgrow you."

37signals has found that there's more people to sell to at the bottom, and when customers need/want more, they're free to find it elsewhere.

For reference, here's our original post on this very topic in June of 2006 when Basecamp was 2.5 years old.


Today Basecamp is 7 years old. Signups are stronger than ever.

Whenever we survey customers asking them what they love most about Basecamp, the top response by a mile is: it's simple, easy, and their co-workers and clients actually use it. It's not multiple choice either - the words "simple" "easy" "intuitive" show up more than any others in the open ended textarea.

We've made Basecamp a lot better over the years. Some people still ask for more. Others say it's too complicated and they wish it was even simpler.

Software development is a challenge. Everyone wants something different. So you do what you can to thread the needle and make as many of the right customers as happy as possible. Not everyone is the right customer.

It sucks to lose a customer because we did something wrong, but it's OK to lose a customer if we just aren't the right fit anymore. People move on from all sorts of things. Clothes, houses, cars, jobs, relationships... Why not software? As circumstances change, one product may not fit someone forever. That's OK as long as it fits plenty of other people at the same time.

Some customers stick with you forever. Others come and go. Many who go come back after trying other tools that promise them more but that no one actually used. In the end, the tool that actually gets used is the one that's the right fit for someone. It's really really hard to get people to actually use things.

We've found that the simplest stuff is what actually gets used. It's why email is still the world's most popular project management tool.

I wrote my answer here: http://bit.ly/iKyBQt

Antennagate and now Jasongate:)) Make sure you don't lose another customer by responding to another customer:)

Ah downvoting... lot of 37signals guys...

Regular HN users also downvote inexplicable comments that add negative value.

This is a really good point. 37signals approach make a lot more sense with that in mind.

Still, it's a painful transition to outgrow a tool you've become dependent on and need additional functionality to do your job.

It's a good point if you don't think that you _can_ develop feature-rich software and still keep the interface nice and easy. It's an excuse if you believe that you _can_...

I think you need to walk the line on this one. That's what we do with Apollo... We managed to include a CRM in there, and still keep the UI very simple.

Tony Mobily http://www.apollohq.com

Change is painful! However, for some, so is learning to ride a bicycle without training wheels. And I say this in all seriousness.

Then perhaps 37signals need to make exporting existing Basecamp data for easy importing into the "next step up" products. As far as I know they only provided a raw XML data dump.


How else would you have it provided to you? That's pretty well the most versatile form of the data you could get, I'd think.

Agreed. It's really up to those other products to develop import routines for BC data. Wordstar didn't develop 'save as MS Word' functionality, but MS did WS imports just fine.

You have it backwards: tools that want to be the next step in the chain should make it trivial to import Basecamp data. 37signals has done their part by making it easy to get data out.

You can also export your data in HTML format so you can still view your projects locally on your own computer:


I explained this exact point to a friend today at lunch. He vehemently disagreed with me.

No matter what the niche, or the market, there is always room for the simplest solution. Always.

That may be, but when the provider of said simple solution is simultaneously ratcheting up the price, people who are using the product will bleed over into other products.

It seems to me that the people who are attracted to the most basic set of features will also have the least amount of pain shifting to a new product while having the lowest price elasticity.

Apple does OK charging high prices for simple products. The target market is prosumers - people who pay a bit more for a product that is (or is marketed as) a joy to use.

Most software engineers are prosumers, so you either target them or the enterprise.

Now, "onion" products are better - they let the user grow from a simple base. But a simple product is a lot easier to make, which frees up resources for marketing or other products.

Apple doesn't typically raise prices - usually same price point and better hardware, more features, etc. The OP was taking issue with coming back to Basecamp and having to pay more than he used to, without getting any new substantive features.

If Apple did this - raised their prices while also taking a 'less is more' approach wrt features, many of their most ardent supporters would eventually defect.

re: basecamp - I knew several orgs in 2008-2009 that had paid versions of it. none of them liked it, but used it because it was something, and they were paying for it so it must be OK. I suspect that over time, BC will bleed enough paying users to the point where they'll reinvigorate the team and start adding more value. Then again, they may just raise the price and ride out their name brand making boatload of money for many more years to come.

Apple products are not simple. Apple products are extremely complex. Their genius is in making extremely complex things usable, so much so that they seem simple.

If I make the world's simplest todo list, I have not made an Apple product. I have made a toy. If I make a phone into a tablet computer that is immediately understandable to a child, I have made an Apple product.

"Apple" = "genius" (OP) + prosumers love Apple (above)

Can't we have a bit of cooling down on this Apple fan thing? The first and last Apple computer I booted myself did crash hard and there was no way to recover, go into BIOS, "hack" it if you will. Ok it was 10 years ago but still, when I see an Iphone owner I ask: show me your files, your music and pic files. Don't you /own/ these?

How can you hack something if you don't have access to its files? When I mount my Ipod, I always wonder why tf they did rename all my music to F01.mp3, F02.mp3, etc. And because of this renaming there is a silly upper limit on the number of tracks I can hold on this said wonderful gadget (I use it because it has a 80G sized belly).

We should be hackers here (and defend the positive meaning of this word given by pg, particularly now). I understand why a CEO would love to use a nice looking toy as their phone, even in a tech startup. But I feel any true hacker should/would feel imprisoned with any device that you cannot mount and scp to, right?

Why would I care about files? I want to see a photo, I want to listen to a song, I want to open a document. I why would I want to hack it? I don't want to hack my car, I want to use it. Ok, I am an iPhone programmer, so I need to see below the surface. But the typical user does not. I am sure you know Clarke's law "Any sufficiently advanced technology is indistinguishable from magic". Do you know what makes it so? Invisible implementation details. File system is an implementation detail. RAM vs. ROM is an implementation detail which exists solely because of the technological limitations. The classics of UX argued about need to hide these details long before iPhone.

Why do you think we see the rise of different "clouds"? Because they hide another layer of details. And yes, they move files you love so much further away from us. For the best, in most cases.

The times when hackers and users were the same people a over. Hackers today must understand, that the users of their products are no longer other hackers (with exceptions, of course) so different rules applies. Normal user will not care about implementation details. They may scare him. What the user want is to pick your product up and just use it. Failing to understand that you fail as a hacker. On the other hand, succeeding to hide implementation from the users is a great hack.

> I why would I want to hack it [document]?

Well, have you never been in need of opening a .doc with a simple text editor like vim and fix some issues? Really?

> The times when hackers and users were the same people a over.

That not my point. I reacted to the false claim that prosumer (ie power users / hackers / hn readers) prefer Apple.

As a matter of fact, most geekest geeks I meet in my job are all having Nexuses. Maybe because I live in China, but it was about the same in France and in Germany.

Interestingly, I find having that low-level access to be more restrictive. Suddenly, a thing that I just want to work (in this case, the music player) requires that I think about how I'm going to put files onto it, what the best way to synchronise those files is going to be, whether or not I have to mount it myself, if I ought to write scripts to manipulate it, the feeling that I need to write scripts to manipulate it, else why have all of these programming skills?

When really what I wanted to do was just listen to music while I'm walking around, and have a limited amount of worrying when it comes to getting music onto the device. (IE, it Just Works territory) My behaviour is now circumscribed in other ways - my attention is split between things I really want to work on and this malaise of I ought to be crafting better tools to work with my music player.

"You can just set up rsync once!" is kind of a non-argument response to this - an Apple gadget, in this case, I set up nonce. Plug it in, it initializes, and syncs everything. Now I can listen to music where-ever I happen to be, and I don't have those circumscriptions to my attention.

If my Apple device annoys me sufficiently that I really want to have that level of control, there's a myriad of devices out there that automount as USB devices, that I could set up all sorts of nifty on-insert rules and script to high-heaven.

"I find having that low-level access to be more restrictive."

This look like an orwellian paradox to me, like "thinking less makes you free". Having low-level access don't mean you /need/ to use this power, it means that you /can/. Not having this power means you gave the keys to someone else. I'd rather keep most of the keys on my side, personaly.

The first time I saw an Ipod, it was at a party, in a friend of mine's hands, looked good, he had nice music, then I asked him to give me a few mp3, and it was "impossible right now". My god...

--> Never bought an Apple product myself (the Ipod I use is my wife's).

Maintaining a competitive edge with software that a small team can produce free of charge is a bit more challenging than a multibillion dollar hardware and software company with factories, no?

I disagree - nobody ever gave up email.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact