back

by raphlinus·9y ago·view on hn ↗
I have a largish open-source portfolio, including a markdown parser, a regex engine, some music synthesis, and some more researchy stuff like a font renderer and a prototype of concurrent text editing using CRDT's. I'm lucky to be working at Google where I get paid 20% time to work on this, but the motivation is definitely not money.

The biggest item in my portfolio is xi-editor, and I confess I'm wrestling with some of the questions raised in this thread. I think it has the potential to be a serious player in the editor space, with extremely high performance goals (including fast startup and low RAM usage) yet a modern feel. It also has a great little open-source community around it who have been contributing significant features.

Yet it's at the point where it's _almost_ done enough to use for day-to-day editing, and I'm hesitating a bit before pushing it over the line. I think I'm scared of having lots of users. It's also the case that I'm very interested in the engine and the core of the UX, but the complete product needs a plugin ecosystem and along with that ways to discover, upgrade, and curate the plugins (including making sure they are trustworthy, lately a fairly significant concern). That's potentially a huge amount of work, and it doesn't really line up with my interests.

I'm wondering if it's possible to focus on the parts I care about and try to foster the community to take care of the rest, but I'm not quite sure how that would work.

If this were a business and I had some way of making a few coins from every user, then my incentives would be lined up to make the best overall product possible, including the less fun parts. But that's off the table; among other things, there are a number of good free editors out there, and the niche for a better but non-free editor is also well occupied.

Maybe the HN crowd has some ideas?

4 comments
Went looking for it, and it's showing up as mainly developed/owned by the Google org on GitHub:

https://github.com/google/xi-editor

Kind of feels a bit weird to be considering if/how/etc to commercialise it personally, when it would potentially be considered Google's property (?). Even though you're clearly the main author as per its commits & README.md. Then again, I have no idea how Google looks upon that kind of thing, so you may be all good.

With the "it'll probably get a million users quickly" thought... hmmm... depends if you're thinking to leverage Google's reach in some way. If so, then yeah it might have a better than even chance to happen. :D

The thoughts on commercialization are a hypothetical. As I said in my original comment, it's off the table (though I gave other reasons than being employed by Google). I'm just saying that if there were a revenue stream, then I'd be incentivized very differently (and more in line with what users need) than purely as a labor-of-love open source project.

And yes, if there were a good reason to, Google could bring considerable resources (including marketing) to bear.

No worries. I somehow didn't pick up on it being a thought experiment. Long day I guess. :)
and I'm hesitating a bit before pushing it over the line. I think I'm scared of having lots of users.

A) What is "lots" of users? (Give me a number)

B) What makes you think you would suddenly have lots?

Cuz most things seem to struggle to get any traction.

A) I'll toss out 1M because you asked for a number.

B) Because if I build out my full ambition, it would be better by most objective metrics (speed, features, integration with IDE-like capabilities) than all the free editors out there, and I think there is a real demand for a better editor.

I know what you mean about the struggle for traction. It is of course possible I'm massively mis-estimating the potential userbase.

It is of course possible I'm massively mis-estimating the potential userbase.

That is largely irrelevant. Do you have a marketing plan? Because "build it and they will come" tends to work poorly.

If you have no marketing plan, your concerns about a sudden influx of 1M users is likely seriously exaggerated.

Does one need a marketing plan if it's not done for money?

While I'd never heard of xi-editor before now, I'm definitely among those who yearn for a "better" editor for (first off) MacOS... and I think there are many others, based on my totally unscientific assessment of the currently popular editors. I'm still on TextMate for a reason. :-)

So my point is, for certain things "build it and they will come" might work fine. If I ever find an editor I like enough to switch to then I will be singing its praises loudly on the Innernets.

I do a lot of blogging on topics where people need info. All the info I put out is available for free. The most popular page on one of my more heavily trafficked sites has around 40k page views.

A million users are unlikely to just magically know your free editor exists. Things do sometimes go viral, but that isn't the norm for how you get to a million users.

Maybe marketing isn't the best term. But, if the OP does not have some idea of how those million users will be attracted, this is probably not a thing they need to be fretting about.

Right, I didn't write clearly. After a couple months of pushing to a 1.0, I'd expect maybe a few thousand users (this seems reasonable to me because it's the number of github stars). That's enough though that I'd need to start worrying about the product rather than just the technology, but that could all be done part-time.

My estimate of a million users is what might happen after another year or two of building out the whole vision, and with organic growth from numerous sub-communities (for example, I'd expect Rust users to be quicker to adopt it because it's in Rust). At that point, it's clear that the project would need at least a full time person to be its maintainer. And I don't see how to sustain that.

Thanks for emphasizing the focus on marketing; it is indeed a knob I can turn.

Thanks for emphasizing the focus on marketing; it is indeed a knob I can turn.

I think my main point is that it is a knob that can be turned in either direction. It seems to actually be easier to dial it down if the attention you are getting is somehow problematic.

Best.

>xi-editor

I read the github readme.md, and I'm wondering.

1. Why should want to switch if I already have a text-editor? 2. Reading through, there seems to be an emphasis on clean data structures and interfaces. Have you taken a look at Kakoune? Because your text editor seems to share quite a few design decisions with it. 3. So, there doesn't seem to be any mention of novel UI/control features even in the front-end. Is that a deliberate choice to emphasize the technical aspects as the selling point (or is it lacking/outdated in documentation), or are there just not any in the first place?

1. If you're using an Electron-based editor, I imagine you might not be entirely happy with performance. I can fix that. 2. Yes, have looked at Kakoune, and even have some plans to emulate its keybindings. It's an interesting project. 3. I'm not sure novelty is the main thing people need. I think we pretty well understand what people need, and there's something to be said for just doing a good job delivering it.
I liked the disclaimer on your github readme. :)