Hacker News Reader

Comment by bengotow | Hacker News Reader

Hmm, I thought Mozilla had pulled financial support from Thunderbird? This would be an exciting new chapter for the product, but the article doesn't address whether it makes financial sense for Mozilla to invest this much in Thunderbird's future.

For most of the last three years I've worked on Nylas Mail (https://github.com/Nylas/Nylas-mail), and it hit 2.0 yesterday with Mac/Win/Linux support. It's entirely open source (GPL), written in JavaScript, syncs mail locally, etc. It essentially -is- what the author of the post is looking to build, except it doesn't look like Thunderbird. Nylas put at least $2M+ worth of engineering time into it, and I imagine ThunderbirdJS would take at least a similar commitment of time and effort. Heck, maybe they could just skin Nylas Mail and call it Thunderbird ;-)

View on HN

Is the "open source" label still comically misleading? [0]

I tried it out, didn't read the fine print, filed an issue, never looked at it again.

Because "It keeps user data local and private." is kind of one of the main benefits.

[0]: https://github.com/nylas/nylas-mail/issues/860

Ahh thanks I need to update that!

Originally Nylas Mail was OSS, but tied to the Nylas Cloud APIs (which did the actual mail processing.)

The cloud strategy was too expensive, and in version 2.0 the entire mail-sync system was moved into the client and open sourced. (https://github.com/nylas/nylas-mail/tree/master/packages/cli...). It's now much more like Thunderbird. There's a bit of cloud processing for snoozing / send later, etc., but it no longer syncs all your email through the cloud.

Is there a way to configure it so it makes zero requests to your servers?

If not (yet), would it at least be technically feasible, and would supporting this be an option for you?

Definitely feasible but it's not our product focus right now. Most of the features that help differentiate Nylas Mail require some cloud component. (snoozing, send later, reminders, open tracking, etc.)

We totally encourage folks to fork the project and take it in whatever direction they want. If you want to chat with other devs in the community, you should join our Slack room: http://slack-invite.nylas.com/

(I work at Nylas)

> (snoozing, send later, reminders, open tracking, etc.)

None of that has any appeal to me, I'd much rather a mail system which doesn't require using someone else's server.

Maybe this isn't a priority for other people, but just putting it out there. This is what's making me not use Nylas today.

He's already given you a good alternative – fork and collaborate with other devs on a version that doesn't touch their server.

Seems sufficient.

I would consider forking a last option, instead of "a good alternative".

A good alternative in my opinion would be just to make some parts optional. Does not sound too hard, to implement (if the codebase is well done)

I remember the last time I looked at Nylas, it was a fancy gmail frontend. Now it appears to require a "Nylas ID" which is not something I need. I don't want my mail being synced to someone else's computer. Where's the "host your own sync system" package?

Honestly when I was reading this post, I was thinking "Why don't they just use Nylas? Nylas has already done this" – it just seems like a perfect case of not-invented-here.

I really don't see the point of Thunderbird reinventing the wheel to essentially build another Nylas.

That was honestly my thoughts too.

I used Thunderbird years ago; now I'm a very happy Nylas user. I don't doubt that the Thunderbird team could spend the next 3 years building a very polished desktop mail app using web technologies, but uh...

...do we need another? And if so, wouldn't it be better to fork Nylas as a base? I mean, they could probably get most of the way their with a couple pull requests to make themes more powerful and a custom theme...

Congrats on your work on Nylas -- 2.0 is shaping up great and is already my primary email client on Linux.

Question: Will the revenue from the Pro product really support development and maintenance of the free product? The free product is perfect for my needs, none of the Pro features are that useful for me (and its too expensive - email accounts from Google or Fastmail are $5/month, Nylas Pro is $12/month on top of that just for the client). The free product, as is today, is easily worth $20-30, IMO.

Additional Comment: https://www.nylas.com/pricing needs to be updated. A lot of the Pro features are now available in the Free 2.0 version.

I suppose they want to use their own UI and maybe link to some services or tools from Mozilla. But they should contact you to unite their effort with your: development time reduced from 3 years to 3 months (that is an approximation)...

At the very least I don't need a "Nylas ID" and Nylas cloud stuff with Thunderbird. Just a client and (my own) IMAP server.

> Heck, maybe they could just skin Nylas Mail and call it Thunderbird ;-)

Skin it or Fork and Skin ;)

That look really good.

Great job.

Reply on news.ycombinator.com