Live data from Hacker News

Mailpile: Let’s take email back

mailpile.is

101–110 of 241 posts

Re: Mailpile: Let’s take email back

#101
post #3

Bjarni here, lead dev. Happy to answer any questions.

Why AGPL rather than a more liberal license like BSD or MIT? Edit: Just to clarify why I'm asking this, the project looks very interesting and I suppose the choice of AGPL wasn't made lightly, as it might slow down adoption by commercial organizations, so I'm interested in the reasoning behind this choice.

In an ideal world BSD/MIT would be the best choice.

But in the current startup world several aynrandian douchebags would take it and make web services or paid apps. Put a lot of money on viral marketing. Pester them with feature requests. And never, ever contribute back to the community or acknowledge it helped their business.

I know a famous BSD developer who switched to *GPL because of this. Also, search for tptacek's explanation. It's counter-intuitive but this is the best license at the moment, mostly for reasons Stallman didn't forsee.

Re: Mailpile: Let’s take email back

#102

Earlier quoted context omitted.

Why AGPL rather than a more liberal license like BSD or MIT? Edit: Just to clarify why I'm asking this, the project looks very interesting and I suppose the choice of AGPL wasn't made lightly, as it might slow down adoption by commercial organizations, so I'm interested in the reasoning behind this choice.

This is something we have debated extensively. Our main goal is to improve privacy and security on-line - giving cloud providers yet another free tool with which to build proprietary services and lock in more users and their data feels like it would be counterproductive. However, we are open to discussing, and if a significant fraction of our backers would prefer a more liberal license we will probably switch.

Please stay AGPL. It's the right choice for these times.

BTW, beware of code ownership of pull requests from corporations.

Re: Mailpile: Let’s take email back

#103
post #94

Earlier quoted context omitted.

I have actually been working on Mailpile "altruistically" for a while, but sorely missed the ability to dedicate myself to it full time. Thus the fundraiser.

Totally get it, but why not share some of the risk with your supporters and meet them halfway? Asking for a full (decent) salary for two people for a year places all of the risk on your supporters and makes you look uninspired. Edit: I can't reply to child, so I'll do it here. It has nothing to do with a free ride. I don't want the product because the solution isn't right for me, but I would much prefer to pay a mont…

$4166/month is meeting people halfway, considering these guys could easily make twice that in salary at a normal job.

Re: Mailpile: Let’s take email back

#104
post #14

I applaud these goals however if we were to ever consider moving from Google Apps (which we would) it would only be for an open source project given that privacy is becoming the more critical issue as the years go on. It would be great to see a subscription based email service such as this where the source is fully available for scrutinizing and patching. Edit: Great, so the source is already available.

You shouldn't trust a cloud store in any case, and you shouldn't need to. Client software, on the other hand, needs to be open.

Re: Mailpile: Let’s take email back

#105
post #84
post #66

I'm in the crowdfunding business, so I feel strange saying this, but I'm really turned off by this model of 'pay us in advance for developing a product that hasn't been market tested or validated,' especially when asking for a sum as large as is $100,000. What happened to building something and selling it? According to the campaign page, the need is so great because: "We're asking for a lot of money, so of course you…

> I'm really turned off by this model of 'pay us in advance for developing a product that hasn't been market tested or validated' Isn't that the whole premise behind a huge chunk of the projects on indiegogo and Kickstarter? "I want to make a cool thing, but I can't afford to quit my day job; please pay me up front"? I mean, I'm truly concerned about online privacy, but there are several barriers standing between me…

Except with VC, the money is given with the expectation of an occasional huge return. With a crowd-funded open source project, one there is no possibility of getting anything more than a program you could get for free if others funded its development. This is clearly a very different situation.

Re: Mailpile: Let’s take email back

#106
post #85
post #66

I'm in the crowdfunding business, so I feel strange saying this, but I'm really turned off by this model of 'pay us in advance for developing a product that hasn't been market tested or validated,' especially when asking for a sum as large as is $100,000. What happened to building something and selling it? According to the campaign page, the need is so great because: "We're asking for a lot of money, so of course you…

There is a thorny issue with kickstarting open-source projects like this. Suppose a third-party developer can make a major contribution to the project, to solve a nasty bug or add a technically challenging feature. How does that developer get paid? How is their contribution valued?

It's not all that thorny really; you don't get any money from your third party patches. The indiegogo funding is for three man years of funded development work split between two men. It's opensource so you are free to help out if you want, just don't expect to be paid.

Re: Mailpile: Let’s take email back

#108
post #101

Earlier quoted context omitted.

Why AGPL rather than a more liberal license like BSD or MIT? Edit: Just to clarify why I'm asking this, the project looks very interesting and I suppose the choice of AGPL wasn't made lightly, as it might slow down adoption by commercial organizations, so I'm interested in the reasoning behind this choice.

In an ideal world BSD/MIT would be the best choice. But in the current startup world several aynrandian douchebags would take it and make web services or paid apps. Put a lot of money on viral marketing. Pester them with feature requests. And never, ever contribute back to the community or acknowledge it helped their business. I know a famous BSD developer who switched to *GPL because of this. Also, search for tptace…

It's counter-intuitive but this is the best license at the moment, mostly for reasons Stallman didn't forsee.

Aren't the reasons you outline in your comment exactly (a subset of) the reasons that Stallman foresaw?

Re: Mailpile: Let’s take email back

#109
this looks awesome, lots of luck.

Am I understanding this correctly that this is basically a selfhosted imap/pop/smtp frontend?

I am curious how you guys would use it, i.e. where would you actually host your emails? Running your own mailserver on a vserver sounds fun and not to expensive but I don't know if I want to maintain something like that in the long run. If something breaks this just sounds like a lot of work.

You guys pay for mail hosting? Use the one that comes with your website domain?

Re: Mailpile: Let’s take email back

#110
post #101

Earlier quoted context omitted.

Why AGPL rather than a more liberal license like BSD or MIT? Edit: Just to clarify why I'm asking this, the project looks very interesting and I suppose the choice of AGPL wasn't made lightly, as it might slow down adoption by commercial organizations, so I'm interested in the reasoning behind this choice.

In an ideal world BSD/MIT would be the best choice. But in the current startup world several aynrandian douchebags would take it and make web services or paid apps. Put a lot of money on viral marketing. Pester them with feature requests. And never, ever contribute back to the community or acknowledge it helped their business. I know a famous BSD developer who switched to *GPL because of this. Also, search for tptace…

what does Ayn Rand have to do with any of this?
Post reply on HN