Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Show HN: fman – file manager for programmers

fman.io

Post reply on HN
Print view View original

81–90 of 216 posts

Re: Show HN: fman – file manager for programmers

#81

Post by mherrmann »

mcgrath_sh wrote: post #79

I think your business model is interesting. However, I'd be careful in letting users decide the direction of the project. It might be okay-ish now if the userbase is small, but when you hit a not unreasonably large number of people (500? 1k?), that can quickly deteriorate. What if the most upvoted suggestion is something that you have no use for? Or if it is something that no future customer would want? Are you going…

That's an interesting point. I've had feature requests I don't agree with. The problem is that one feature may make one person happy, but (say) complicate the UI and thus make another person unhappy. It's a balance between my vision and what people want.

Re: Show HN: fman – file manager for programmers

#82

Post by jmcdiesel »

mherrmann wrote: post #80
jmcdiesel wrote: post #75

Earlier quoted context omitted.

How are you "sold out" of a digital license? Sounds like a very Nintendo-esque intential shortage to drive demand. Not saying its true, but it was my first thought, as well as 5 of the people in the slack room i shared the link in... all of which went from "ok, lets give this guy money" to "... nah... this doesnt feel right, now..."

It's a good question - I already explained this here in a comment: https://news.ycombinator.com/item?id=13777696

You really haven't answered it, though. It's very nonsensical. If you want to see "who's in" and develop for your paying customers, then should you not let people become paying customers so you can develop for them?

Your logic seems to work if you were not allowing trials, but only paying... THEN you would have only paying customers... its your gig, but I think its been a while since i've seen something this strange in software... and its telling that you're having to post the link to your answer so often in these comments. It simply makes no sense, and seems to be serving to drive people away or make them doubt your intentions.

Re: Show HN: fman – file manager for programmers

#83

Post by sdegutis »

mherrmann wrote: post #81
mcgrath_sh wrote: post #79

I think your business model is interesting. However, I'd be careful in letting users decide the direction of the project. It might be okay-ish now if the userbase is small, but when you hit a not unreasonably large number of people (500? 1k?), that can quickly deteriorate. What if the most upvoted suggestion is something that you have no use for? Or if it is something that no future customer would want? Are you going…

That's an interesting point. I've had feature requests I don't agree with. The problem is that one feature may make one person happy, but (say) complicate the UI and thus make another person unhappy. It's a balance between my vision and what people want.

It'll be interesting to see how that plays out.

Btw why are the Linux and Mac versions both 21mb but the Windows version is only 1mb? Magic? I'm gonna assume magic.

Also: it doesn't seem to use Electron! Cool! But wait, it uses Python + Qt instead?

Re: Show HN: fman – file manager for programmers

#84

Post by Jare »

mherrmann wrote: post #63
zerr wrote: post #62

Earlier quoted context omitted.

Is it electron based?

If you mean fman then no: I thought about this long and hard but startup speed is crucial and Electron just doesn't cut it. It's based on PyQt. I wrote a blog post about this: https://fman.io/blog/picking-technologies-for-a-desktop-app-...

I see the Windows version is a 1MB download and the Mac & Linux are 21MB, so I was surprised to ready Python and Qt are involved. I take it the Windows setup is a downloader?

Re: Show HN: fman – file manager for programmers

#86

Post by mcgrath_sh »

mherrmann wrote: post #81
mcgrath_sh wrote: post #79

I think your business model is interesting. However, I'd be careful in letting users decide the direction of the project. It might be okay-ish now if the userbase is small, but when you hit a not unreasonably large number of people (500? 1k?), that can quickly deteriorate. What if the most upvoted suggestion is something that you have no use for? Or if it is something that no future customer would want? Are you going…

That's an interesting point. I've had feature requests I don't agree with. The problem is that one feature may make one person happy, but (say) complicate the UI and thus make another person unhappy. It's a balance between my vision and what people want.

Exactly! Granted, this is a very early stage of development, so it may not be as big of a deal.

Another thing to keep in the back of your mind (if you don't mind more unsolicited advice) is that you are building something that is the base for plugins. For it to be as successful as Sublime is, the core has to be rock solid and be made up of core features only. Be careful with feature creep. And consider putting things that you may not feel are part of that core into plugins initially (think vintage in Sublime). If they are, then migrate them over.

(I promise, no more telling you how to oversee your project!)

Re: Show HN: fman – file manager for programmers

#87

Post by mherrmann »

sdegutis wrote: post #83
mherrmann wrote: post #81

Earlier quoted context omitted.

That's an interesting point. I've had feature requests I don't agree with. The problem is that one feature may make one person happy, but (say) complicate the UI and thus make another person unhappy. It's a balance between my vision and what people want.

It'll be interesting to see how that plays out. Btw why are the Linux and Mac versions both 21mb but the Windows version is only 1mb? Magic? I'm gonna assume magic. Also: it doesn't seem to use Electron! Cool! But wait, it uses Python + Qt instead?

When there's doubt about feature requests, I side with conservative choices because I hate bloated software.

The Windows version is an online installer that always fetches the latest version from the update server. It's based on Google Omaha, the same technology that also Chrome uses for auto-updating. I wrote a blog post about this [1].

It uses PyQt instead of Electron. I was very tempted to use Electron, but its startup time is prohibitive. I wrote a blog post about this as well [2].

[1]: https://fman.io/blog/google-omaha-tutorial/

[2]: https://fman.io/blog/picking-technologies-for-a-desktop-app-...

Re: Show HN: fman – file manager for programmers

#88

Post by mherrmann »

mcgrath_sh wrote: post #86
mherrmann wrote: post #81

Earlier quoted context omitted.

That's an interesting point. I've had feature requests I don't agree with. The problem is that one feature may make one person happy, but (say) complicate the UI and thus make another person unhappy. It's a balance between my vision and what people want.

Exactly! Granted, this is a very early stage of development, so it may not be as big of a deal. Another thing to keep in the back of your mind (if you don't mind more unsolicited advice) is that you are building something that is the base for plugins. For it to be as successful as Sublime is, the core has to be rock solid and be made up of core features only. Be careful with feature creep. And consider putting things…

Thank you for the advice :) Actually, all the basic functionality - copying/moving files, the Cmd/Ctrl+P "GoTo on Steroids" feature etc, are implemented in a plugin. I do this to make sure that fman's (Python) plugin API is powerful enough that plugin developers can do just as much as me.

Re: Show HN: fman – file manager for programmers

#89

Post by jmcdiesel »

fernandotakai wrote: post #85

> don't be a free user! sure, going to buy it > i'm sorry, we're sold out :-( 100 licenses were made available on March 1, and all are gone. guess i will be a free user :/

Never seen a developer both talk about the rationale of not being OSS to cover costs, then decline people wanting to give him money...
Post reply on HN