Why 37signals is falling into obsolescence | Hacker News Reader

Why 37signals is falling into obsolescence

seancoleman.net

46 pointsbdclimber1449 comments

While a mobile app can certainly deliver a rich user experience, the availability of a standard, universally supported platform (read: a browser) is a more sustainable approach to application deployment.

We've been down this road before. In the pre-browser era, desktop application were king, but at a high cost to IT departments as well as end-users. The same game is being played out on mobile platforms now.

From a business perspective, do you really want to support multiple development efforts for each and every mobile platform? Will it be worth it in the long run?

The advantages of the mobile app platform must be balanced with the cost and maintenance overhead as well. Give me a great browser platform, and I'll gladly take its shortcomings over the long-term sustainment costs of a dedicated mobile app.

As a customer though, what would you prefer? That was my main point is that the customer demand is really heavy on native apps.

Evernote is an example of a successful startup that has native apps for every major platform, and they consider it a strategic decision.

Only 37Signals is a B2B company that has a lot of customers that are developers. They have an API. Their source of income is their subscription and not apps for native platforms.

In other words, many of their customers who want a native app can scratch their own itch pretty easily, and 37Signals can cut down their own development efforts and focus on making their platform better.

The reason Evernote has apps for every platform is simple: that's the only thing they have going for them. Their interface isn't the best, and there are tons of other ways to save notes. They just happen to have really good cross-device integration. Other note-taking tools have users for different reasons.

I agree with 37Signals. App Stores are an obsolete concept that are only making a comeback because we have a new wave of devices without a solid set of mature cross-platform development tools.

EDIT: In other words, 37Signals wants to differentiate themselves from other applications in more unique ways than just "works natively on every device."

Honestly, as a customer I'm getting tired of having to install apps for everything I use, I'm getting tired of something not having the app on my Droid because it's only for iPhone and so on...

I think the HTML5/JS approach still has a way to go before the user experience will really catch up to native apps, and for some apps it will never happen. But the same could be said for desktop apps vs webapps and we all know how that turned out...

As a customer I prefer an application that works on whichever device I feel like using this week.

Having only a mobile web version isn't necessarily obsolete.

The current demand for native apps on mobile devices stems from the fact that it is easier to get high quality user experience on the native apps than on mobile web browsers. This was true of desktop apps, too, 5-10 years ago. However, web-apps offer a host of other advantages and so long the UX can be brought up to a decent quality, I'd take a mobile web-app over a native app that I have to go find and install for the same reasons I prefer web apps to desktop apps (both to use and to create).

So, 37s's decision to do a mobile web-app is either obsolete or visionary. I'd bet on the latter.

It always amazes me that people don't see we're repeating desktop history.

Look back to when desktop apps were the norm. People eventually realized it was counter intuitive to write the same program 5 times for 5 different platforms so they started moving to frameworks (Java/.Net). Then as the web gained the power to host comparable apps people started moving to the web.

It was a natural progression of societal knowledge based on trial and error.

So why are we now making the same mistakes on mobile platforms? Especially since web apps can already match the functionality needed by 99% of the applications out there? Seems a little silly

That 1% still covers:

Access to the device's GPS. Access to the devices Accelerometer. Hardware Accelerated 3d Graphics (Although this is possible with WebGL, I would wager the tools are more focused towards native clients) Access to the device's Camera / Film Roll

You can access the GPS information with Javascript. http://diveintohtml5.org/geolocation.html

Ugh, I could have read something else but I read that instead.

linkbait headlines anyone? Next, learn why some blogger is quitting javascript forever and switching to Erlang: bit.ly/hTNYCW

Come on. It'd only be linkbait if the meat of the article didn't address the title. I thought his title was very relevant to the content.

Thanks, my intention was definitely not to create a linkbait title.

Maybe I'm missing the point, but aren't there tons of 3rd party clients available as well as their own Highrise client?

http://basecamphq.com/extras

I'm not fan of Basecamp, but I think they've got a good ecosystem going.

They have their Highrise client that they hired a contractor to develop on iOS, and decided that it wasn't worth repeating for their other products.

He's right. Many people bet on the web, there is a lot of legacy on the web, and the web is obviously easier for developers to make.

But the customers want native apps (as thick clients to internet services). Native apps work better. They are faster and have more features.

Companies that have so much invested in the web, and choose to stick with this type of technology will find that they were wrong.

I think you are confusing the average smartphone user with the average Basecamp user. The former is interested in shiny apps that make them feel like Tom Cruise in Minority Report. The latter wants something that is fast, easy to use, and saves time.

37 Signals has made a pile of money by ignoring all the cw about what customers want. I wouldn't write them off just yet.

Native apps are faster, easier to use and save time. Web apps are easier to develop for them, that's why they chose them.

This is the argument everyone pushes, but it's not true. Yes, JavaScript is slower than Objective-C. No, it doesn't matter in most cases. Speed of almost all consumer apps is measured in user time, not in ms, and you can achieve top notch perceived speed in a web app for most of the types of apps on the app store.

The problem is not javascript, the problem is server round-trips. Local HTML5 is practically indistinguishable from a native client - but that's not what they are doing. They are making a proper server based system.

All of their products are web apps. Those round-trips are getting made regardless.

You don't think the companies using the shiniest new web apps aren't using the shiniest new phones?

Basecamp is not "shiny". It's functional, fast, and has a very small learning curve. That is why it is such a roaring success in the business community. People in business don't have time to decipher your chrome-laden, non-HIG iPhone app. That is also why Windows XP and BlackBerry still retain a lot of popularity in that community, despite being based on Precambrian technology.

It is the poster child of Web 2.0 services and Ruby on Rails. Has the world really moved on that fast that this is old hat now?

A lot of desktop computer based software migrated to the web where it ran on the server. Made things simpler and avoided "updates downloads" since the updated could be pushed to the code running on the server. With the different platforms, and so many versions of Andriod floating around, doesn't what 37 signals is doing, make things that much simpler? Basecamp is about online collaboration. How could a "native app" do this without the internet?

I think I disagree with the premise, which is that 37Signals is making some sort of strategic mistake by not implementing native apps. My perspective is that back in the '90s, I wrote internal desktop apps for corporations when the web came along, setting back the user experience a decade (at least). Web apps have only recently begun to offer the kind of user experience that desktop apps provide. Because of this, I was surprised when web apps won, and won so quickly and thoroughly. But it turns out that the user's experience - although important, and especially so for consumer apps - is only one of many factors that drive technology adoption, even in the consumer market.

Certainly 37Signals might pick up some users with smoother client apps, but at least for the time being, they have to balance that against the cost of developing those apps. My guess is that most users will care about clunky HTML vs. smooth native apps about as much as they cared about clunky HTML vs. smooth native apps back in '98. Which is to say, hardly at all.

37signals has an extensive API built around all their apps with tonnes of 3rd-party native apps for desktop and mobile. If there isn't a native app for your platform, go write one and stop complaining.

And arguing that a web app isn't useful as _a freaking web app_ on your mobile is a little wild to me. There may be deficiencies in areas like file uploads on mobiles, but largely, you interact with Basecamp through a browser on your desktop. Why doesn't a mobile work well? (I've used many iOS Basecamp clients and none were better than the mobile version launched today by 37s).

I love native apps too, and I'm glad that 37signals encourages their existence by providing a solid, well-documented API. And while they do ship some native apps, I totally buy them building really awesome web apps on the desktop, and now on mobiles.

It's frustrating to me when people say something along the lines of "go write one and stop complaining." It's hardly a useful retort. I think his main thesis is an important one to take note of. One that anyone who runs a software company should consider.

I can't seem to find OrangeSlyce (the author's company) on the iOS App Store. Maybe his company has been obsolete since day one? Our little company doesn't have a native iOS app either. We do make sure our web app works in mobile WebKit based browsers though. Are we obsolete already?

The relevance of a native app is strictly dependent on your users' demands. If I'm going to make an investment in a native app for iOS, Android, Blackberry, etc, there had better be a business case. How many users would sign up for my service based on the fact that I have a native mobile client? What percentage of my users are even using a mobile device to access my service?

It's simply not rational to look out across the tech landscape and declare every company without a native mobile app to be "obsolete".

Our companies, at least OrangeSlyce (far from it), are definitely not at the scale of 37signals, so I would argue its a different set of circumstances.

You're definitely right that its not rational to declare every no-mobile-app company obsolete. My main point is that a company who resolves to a certain technology (web apps in this case) just because they don't know another technology (iOS, Android) isn't adapting to market changes.

I see no reason why anything in the post would apply to OrangeSlyce, based on the OrangeSlyce website. You're comparing a graphic design company to a product company. Graphic design does not go on the App Store — products do.

Relatedly he's not declaring every tech company without a native mobile app to be "obsolete" — he's declaring non-native mobile apps to be obsolete.

I can give you at least one reason:

Paragraph four, in bold: "In order to stay competitive, web companies now must provide for all devices."

An expanded quote from that paragraph: "If I’m walking down a hallway, I’ll be swyping on my Nexus S. When I get to my desk, I’ll switch to my MacBook Pro, and when I jump on a long flight, I’ll be typing on my iPad. I expect demand the absolute best user experience on each of these devices, and vote with my wallet for the company that will provide this."

The goal of OrangeSlyce appears to be connecting designers and customers. It's inherently social. Why wouldn't I want to have ready access to browse projects (as a student) or monitor my listing (as a project stakeholder)? I'm not trying to be a prick here, I'm just encouraging everyone to step outside this situation and get objective for a moment. We didn't go with Basecamp for our project management needs, so I'm not in this to stick up for 37 Signals or even my own opinions on their product.

I'm prepared to cede the fact that the author's intention was to address 37 Signals and similar companies, but I really don't understand why I'm being drilled for taking away what I thought was a pretty clear point.

His title is clearly calling out 37 Signals, not every company. Are you a bit defensive? 37 Signals has thousands of users, maybe even more. It's my understanding that there is constant demand for a native mobile app.

If your business doesn't need a native app, then don't create one. It's really that simple.

So assuming we're talking about 37 Signals only, I'd still offer the rebuttal that claiming obsolescence because of a lack of native app is a stretch. It's hyperbole. Yes, there is a vocal community of users who want a native app, but what are the costs associated with building the native app? Would 37 Signals see any substantial benefit by releasing a native app? What percentage of their user base are dissatisfied with the mobile web app?

There are a vocal community of users that want Flash on their iPad/iPhone/iPod Touch. There are a vocal community of users who want a Facebook iPad app. Apple and Facebook aren't on the path to obsolescence because they don't bend to the whim of their users.

In order for something to be obsolete, it must be operable, but unused. To be on the path to obsolescence, one would expect a decline in use. That's pretty hard to prove in the case of 37 Signals, because they don't publish user-base numbers. Or maybe they do. Does anyone have a source for this information?

You've presented a very good rebuttal that's hard to disagree with.

I could see why you'd think I'm being defensive, given all the 37Signals fanboys out there, but I am not among them. I respect their work, but I disagree with a lot of their ideas about running a company.

I am just struggling to find the point of the article. If he's simply calling out 37Signals, then more power to him. Here I was trying to find something to take away from the article.

His main point is that 37signals doesn't have a native app for Android/iOS but should because of the increasing number of users that own these devices. His secondary point is that a mobile phone web app simply isn't enough. Without rereading the article, he backs this up with an example of low internet signal in the bathroom and how a native app could remedy such a matter. He also mentions an airplane.

That was my simple summary. My takeaway is that software entrepreneurs should consider such issues. As he is right. Mobile web apps aren't enough. The UI paradigm between iOS and Android alone is enough to demonstrate this. I'm going to write a blog post about this tomorrow. Interesting topic that coincidentally I was discussing last night.

37signals provides APIs for its apps and there are many 3rd party native apps available based on them. For example Propane - a native Mac Campfire app is very popular.

If the argument is "they are obsolete because they don't do native mobile", that is a poor argument to make. Where is the data to support it? It's simply a theory wrapped up as a conclusion.

The more important question is, "are they positioned well for the future?" That is really what matters. Don't focus on a particular type of technology as the end-all-be-all of development.

Using the cache manifest, you can still run your web apps offline...

I think he makes a very lucid argument ...

"The fact that 37signals is unwilling to adapt to a changing market (i.e. demand for native apps) shows their archaic mindset and lack of innovation"

Mobile Development seems to be the next frontier in the evolution of the interwebs, and they could very well be betting on the wrong horse.

I think apps are the flavor of the day, but the web will ultimately win.

Agree. Mediocrity always wins.

No. Better economics wins:

Web sites are cheaper to maintain upgrade, and add features to than native applications, which means that for the same price, you can provide more to your users. Think how much simpler it is to provide a web app that works on iPhone, Android, Blackberry and Symbian, compared with building it natively for each one.

Most (non-game) native apps are just refined clients for web services.

They won't go away any more than PCs went away when we all moved from PCs back to the network the last time.

I mention that companies must provide for distributed client access. There are other ways of doing this than developing all clients in-house. Twitter has demonstrated this excellently through their fantastic API. I think a key for startups to focus on is API first, client acces second.

I couldn't vouch for its speed or coverage, but Basecamp has an API which several existing third party native apps use...

I've used a lot of those "extras" but they are all very limited and I think mostly because of their API.

Is Google also obsolete?

Reply on news.ycombinator.com