Mailchimp Blog

In these cases it's really helpful to recognize the distinction between the Mozilla Foundation (MoFo), the Mozilla Corporation (MoCo), and the broader, open source Mozilla project that binds everyone together.

The non-profit Mozilla Foundation exists to support the project, and does truly wonderful, pure work on keeping the Web open. Advocacy, education, grants, etc. They also own the for-profit Mozilla Corporation, which is in trenches directly competing with other browser vendors to ensure that the project has sufficient influence to fulfill its mission.

Persona was housed in MoCo, which afforded it greater resources at the cost of greater organizational risk.

> Either Mozilla is full of it when they say they fight for the free web, or they massively misunderstand how important it is to get rid of proprietary authentication.

The Mozilla project is too small to fight effectively on every front; we have to focus on areas of greatest urgency and where we have the greatest leverage. In the instance of Persona, two major things changed: (1) Persona became less urgent, and (2) we discovered that we couldn't credibly leverage Firefox to boost Persona adoption.

To the first point, back in 2012 we were seeing huge startups, including Spotify and Pinterest, run without offering any means of authentication besides Facebook Connect. The lack of traditional authentication on these sites looked like an ill omen for the freedom of the future Web, but that didn't come to pass. Both…

I think you go freemium if the market is large enough to support the idea that if you just convert a small percentage of customers, you'll still be a successful company. If the answer is yes it's large enough, then whatever strategy gets your software in the hands of the most number of users should be the path you follow.

Now, as a startup, I wouldn't start by competing head to head with OpenTable's salesforce going after their customers. That sounds expensive. Also, getting people to switch providers is always way more difficult. Plus, I think people value OT not because of their features but because of the network effect they bring. Don't get me wrong, you'll definitely still be going one-by-one and converting users, but I don't think you should start with those using your competition.

Instead start with new restaurants or those not using anything. To them, you'll feel like a miracle.

The reason you want to maximize distribution of software and engagement (having something that even the free plan is worth talking to people about) is because going freemium is not unlike investing in startups or building up a portfolio.

You invest in your customers in mass (with free features not money) and you hope a small grouping of them will make back that investment many fold.

Like investing, freemium works best as a long term strategy. So when you look at your upgrade triggers, be sure not to cripple that free plan so much that you're optimizing for conversion rate for users as…

As a "user", I love logging in with social media. How awesome is it that I can just click a button and be logged in instantly? No worrying about setting up a password, at least, for most websites. I used to not trust it, but after becoming a developer and learning: it is just grabbing an email or a name is not too bad. Though sometimes, it may grab more than that, such as your friends list, etc. Somehow, that became okay and no one cares.

As a "developer", things are quite different. There are a few libraries of code on Github and you can find examples across the Internet. For the most part, these examples do work after you have set up some API backend on the social media platforms themselves.

Code would almost seem straightforward. However, it is not always the case, especially, when a social media platform is ever-changing. For example, last week, someone released a new product on Hacker News and went about his way, only to realize, that people were attempting to login to Facebook, which recently upgraded from SDK 4 to SDK 5. Massive changes occurred. How embarrassing to go live, only for one of your social media buttons not to work? It's not his fault, but it's Facebook who updated their code so older versions no longer work. Most developers are trying to focus on their own code and make sure it works. Now they have to worry about potential users being unable to access the front door?

When I went to update Facebook SDK 4 to SDK 5 on one of my own platforms, using…

Wow, and on top of everything else, sparkpost just upgraded their free tier from 10k/mo to 100k/mo sends. That's pretty huge, as 100k/month gets you a lot of runway before you need to really start paying for things. Plus, the CEO promises to keep honoring the 100k threshold for anybody who signs up under that plan, even if they have to lower it again in the future. [0]. I'll have to check them out.

Of course my real problem at this moment is Mandrill just plain works already, and I'm not sure if it is worth spending engineering time shifting to a different provider instead of just paying the premium. TBH, the biggest issue is the feeling of Impending Doom I get from the mandrill discussion. [1] seems to indicate that the transactional email of mandrill really just doesn't fit in the corporate culture of mailchimp. So it's not just the change in pricing that worries me, I'm concerned about a future where there's a lack of innovation and support in the product itself, and I'm still locked in to paying more for less because I don't want to spend the time moving providers. Sunk costs and all that.

0: https://www.sparkpost.com/blog/my-promise-to-developers-sparkpost-pricing
1: http://blog.mailchimp.com/important-changes-to-mandrill/

The latter is a valid question. MailChimp had a post about a year ago [1] where they found that social logins had some negative components. Specifically, that they overall had a lower login rate than the standard username/password. This isn't quite the same, but it IS extra effort for legitimate users to log in.

Also, every new login method requires user training. What does the user education process look like if you had to train and support them on (up to) seven different authentication methods? If a user gets used to one of them (say, proximity), how smoothly can you keep them educated that if they are in the subway, GPS won't work, so they need to try /method X/, then /method Y/, etc. That may be worth the effort to some services where security is absolutely paramount; hopefully anyone using this service builds in some heavy analytics to find out!

[1] http://blog.mailchimp.com/social-login-buttons-arent-worth-it/

If you want to read it, the work-around while their blog is down is hop over to Google Reader, click Add Subscription, and paste in the URL of their blog (http://blog.mailchimp.com). This is my go-to work around for accessing blogs which were killed by traffic.

But what is the work around for writing blog posts such that they don't get killed by routine events such as showing up on the HN front page? Turn off Apache KeepAlive. Apache KeepAlive is to dead blogs as smoking is to lung cancer: not only does A certainly lead to B if you give it enough time, the existence of B strongly implies A!

For more on this topic:
http://www.kalzumeus.com/2010/06/19/running-apache-on-a-memory-constrained-vps/