Hacker News Reader

Comment by teleclimber | Hacker News Reader

Is your product open source? Because for this kind of thing it's really hard to trust unless it's open and self-hostable.

Even if you say you don't need to store emails, there is still a chance of leaks via logging systems or simply programmer errors, or some caching layer. It's likely your security protocols aren't as robust as Google's. (You mention "https" as the primary reason this is secure. Much wow.)

Given the importance of email there is no way I could trust another closed system with my email. I'm sorry if that sounds harsh, I also want to see alternative email frontends because gmail isn't cutting it.

View on HN

It's not open source at the moment no.

You make a very good point, and I see where you are coming from.

I personally have a hard time trusting other products with my data too. That's why I'm trying to be as transparent as possible with DarwinMail.

If I were to make the product open source, would you consider using it then?

Yes, if you make it Open Source and possible to self-host it goes from DOA to definitely possible. Would have to explore more, like how it's written, the community around it, etc... before I really commit.

Hopefully you can still make a living. I suspect "open source + commercially hosted" is a good model for something like this?

You are asking him to do a ton of work and commit to never charging you for anything on the off chance that you might condescend to use his software. It seems like it would be easier for you to just not evaluate new mail clients.

It is so kind of you to defend my position.

Thank you so much. I'm so happy that you understand as it can be difficult for a creator to explain such things :)

I really appreciate it!

I'm just suggesting that a service that touches email would benefit from being easier to trust. I never said he should commit to never making money or anything close to that. Like you literally invented that.

In fact I suggested he look into the open source + hosted business model. Ghost seems to be doing alright with it:

https://blog.ghost.org/5/

> It seems like it would be easier for you to just not evaluate new mail clients.

There are plenty of OSS mail clients. I was just looking at RoundCube the other day. It's actually pretty slick as far as traditional email clients go.

https://roundcube.net/

You have certainly given me plenty to think about.

I thank you :) I love receiving new 'outside of the box' ideas that will increase DarwinMail's appeal :)

You said “Yes, if you make it Open Source and possible to self-host it goes from DOA to definitely possible.[...] Hopefully you can still make a living. I suspect "open source + commercially hosted" is a good model for something like this?”

That means that you would not consider using by it unless you could use it for free, presuming you wouldn’t qualify for a commercial license. I only summarized what you said.

It can be difficult to make a living from your projects as an indie developer.

Thank you for understanding :)

I see the merits of open sourcing DarwinMail. It would open up the platform to develop on it from people all over the world. They would fix bugs for themselves and others and perhaps write mini-features. However, if I open source this code, I may never be able to make any profit on this tool I have spent so much time building :( How will I be able to pay my bills?

> How will I be able to pay my bills?

That's why I mentioned the OSS + hosted version. It would allow to be more open while still making money on the hosted version. Those who are very concerned about privacy can self-host, and those who think self-hosting is too hard (that's many people) can pay to use your hosted version.

See Ghost as a success story that works that way:

https://blog.ghost.org/5/

Thank you for the follow-up.

That's quite an interesting concept! It would alleviate users concerns :)

I like it!

Ghost has done very well in recent years. I will read up on this.

That was my biggest concern as well. I am sure you do not do that, it it would have been so easy to funnel my data (or metadata) to your servers.

Open-sourcing and authenticating the client code would resolve this issue for me.

That is an interesting point and I very much appreciate your opinion.

I will think about it :)

Reply on news.ycombinator.com