5 ms·

Rationale, if you went to the comments first, same as me: https://github.com/LibreSprite/LibreSprite/issues/1 https://github.com/LibreSprite/LibreSprite/issues/

by akavel 8y ago

Rationale, if you went to the comments first, same as me: https://github.com/LibreSprite/LibreSprite/issues/1 https://github.com/LibreSprite/LibreSprite/issues/1 "Since https://github.com/aseprite/aseprite https://github.com/aseprite/aseprite went proprietary on Aug 26th, 2016 (aseprite/aseprite@5ecc356), this fork and corresponding GitHub organization were created to preserve the last GPLv2 version of the code. Our reasons for doing so are multiple and likely different for all fork initiators, but a strong belief in free software is definitely to name, and as the GPLv2 license allows us to, we want to safeguard this nice open source piece of software." - also included in the Readme.

mrspeaker 8y ago
I love open source and I wish it was still GPLv2, but am also a huge fan of Aseprite. It was the best $15 I've spent on software. Every time I open it up there is some update that I'm excited about (and absolutely tonnes of new stuff since 2016 [edit: "1274 commits behind aseprite:master" to be precise]). Plus, the interface makes me smile every time. I'm certainly not against LibreSprite - I'm keen to see where they take it - but just saying, "hats off" to Aseprite!
- pjmlp 8y ago
  This is why it is so hard to make money on desktop FOSS software. As soon as the developer asks to be supported, the run to fork and keep using it for free starts. Then people wonder why most desktop related companies focus on Apple and Microsoft platforms.
  - gman83 8y ago
    "Asks to be supported" is different from "changes the license from FOSS to proprietary". Sure, they can do that, but don't complain when someone forks your software then.
    - pjmlp 8y ago
      Because "Asks to be supported" usually boils down to a few donations and patreon accounts that hardly pay the bills for most developers, unless they have other sources of income.
      - akavel 8y ago
        EDIT: The author's post from 2016.09 outlining the background of his decision: http://dev.aseprite.org/2016/09/01/new-source-code-license/ http://dev.aseprite.org/2016/09/01/new-source-code-license/ Super recommended reading! --------- It's not clear to me if you're aware, but aseprite's author actually had a very interesting approach to monetization already live for some time, AFAIK. Namely, code was GPL, but official Windows binary builds were paid-only. I found it ingenious, though from this change I assume it was deemed not good enough. Curious what were the numbers. As a side note, the original ASEprite (Allegro Sprite Editor) was what made me first discover the Lua language back in the early 2000s. For which I'm eternally grateful :)
        Latty 8y ago
        XChat was using this model for as long as I can remember seeing it.
        Cthulhu_ 8y ago
        And XChat had a fork that was distributed for free, meaning few people would actually pay for it. (anecdotal)
        Jach 8y ago
        Presumably the binaries were listed on the torrent networks, also meaning few people would actually pay for it (anecdotal). Selling end-user software is hard. The license itself probably won't make or break you.
      - PyroLagus 8y ago
  - misterhtmlcss 8y ago
    I don't believe this is true. Feels like confirmation bias a bit. Lots of companies or people make money on their open source software. Vue.js, Red Hat, etc. That's just the top of mind and I'm sure there are more. It looks like lots because you count each failure as confirming your bias, but if you correlate that against over 80% of small businesses failing then it's not a valid assumption unless you have data showing they fail at a higher rate. Don't get me wrong this is a shitty situation for sure, but it's always sad when an entrepreneur of any stripe fails. As an aside I think it's a more valuable conversation in general to look at the business models available to someone approaching this model of development and for us to make suggestions on how he could have built a successful product in this way and succeeded. My thoughts as a total outsider on this issue are going to be more general, but since it's my suggestion to do this I'll start. 1. I believe that when someone creates software in a broad usage category then consulting and especially training must be central to any business goals if you wish for similar projects to succeed. 2. Support contracts for large orgs are also very important because companies need security whether software is free or not. Without support contracts it's likely you won't get the kind of benefactors that are required to be financially sustainable. 3. Marketplaces seem to be in vogue as are extensions. These often work well for one very simple reason; people are paying either directly or indirectly (via their marketing of their extension) to promote your product/service. Nothing says win like others promoting your stuff for you. 4. Build a currated network of developers and promote them. Talk about them to others. Helping them is helping yourself. The more people or orgs investing in your marketplace developers the more stickiness a product will have. 5. Talk about the product everywhere. Conferences. Even start a conference around your vertical. Invite your dev circle to speak at the events. If nothing else it'll boost their careers. Anyway that's my off the back of the napkin 'just saying' for today. I hope it helps someone who's interested in following this gentleman to Foss joy and success. :)
    - pjmlp 8y ago
      The examples you cite work in enterprise space, not when selling desktop applications to regular consumers. Red-Hat explicitly moved away from desktop because of it. Even Ubuntu has given up and decided to pivot into IoT and server. Average steet consumer does not care about consulting, and if some kind of training is needed they ask around to friends or local teaching associations. The original author sees zero from those teachings and everyone in the business knows that desktop software books are more about building CV than actually living from them.
  - pcwalton 8y ago
    It's hard to make money by selling desktop software, period. In 1995 it was a reasonable model, but there's a reason why everyone goes SaaS in 2018. The market isn't growing, and consumers aren't used to paying for typical desktop software anymore (with Mac a possible exception).
    - pjmlp 8y ago
      And SaaS is anything but open source most of the time, specially in the consumer space. We are back to X-Windows applications on dumb terminals, just prettier.

by akavel 8y ago

mrspeaker 8y ago

I love open source and I wish it was still GPLv2, but am also a huge fan of Aseprite. It was the best $15 I've spent on software. Every time I open it up there is some update that I'm excited about (and absolutely tonnes of new stuff since 2016 [edit: "1274 commits behind aseprite:master" to be precise]). Plus, the interface makes me smile every time. I'm certainly not against LibreSprite - I'm keen to see where they take it - but just saying, "hats off" to Aseprite!

pjmlp 8y ago
This is why it is so hard to make money on desktop FOSS software. As soon as the developer asks to be supported, the run to fork and keep using it for free starts. Then people wonder why most desktop related companies focus on Apple and Microsoft platforms.
- gman83 8y ago
  "Asks to be supported" is different from "changes the license from FOSS to proprietary". Sure, they can do that, but don't complain when someone forks your software then.
  - pjmlp 8y ago
    Because "Asks to be supported" usually boils down to a few donations and patreon accounts that hardly pay the bills for most developers, unless they have other sources of income.
    - akavel 8y ago
      EDIT: The author's post from 2016.09 outlining the background of his decision: http://dev.aseprite.org/2016/09/01/new-source-code-license/ http://dev.aseprite.org/2016/09/01/new-source-code-license/ Super recommended reading! --------- It's not clear to me if you're aware, but aseprite's author actually had a very interesting approach to monetization already live for some time, AFAIK. Namely, code was GPL, but official Windows binary builds were paid-only. I found it ingenious, though from this change I assume it was deemed not good enough. Curious what were the numbers. As a side note, the original ASEprite (Allegro Sprite Editor) was what made me first discover the Lua language back in the early 2000s. For which I'm eternally grateful :)
      - Latty 8y ago
        XChat was using this model for as long as I can remember seeing it.
        Cthulhu_ 8y ago
        And XChat had a fork that was distributed for free, meaning few people would actually pay for it. (anecdotal)
        Jach 8y ago
        Presumably the binaries were listed on the torrent networks, also meaning few people would actually pay for it (anecdotal). Selling end-user software is hard. The license itself probably won't make or break you.
    - PyroLagus 8y ago
- misterhtmlcss 8y ago
  I don't believe this is true. Feels like confirmation bias a bit. Lots of companies or people make money on their open source software. Vue.js, Red Hat, etc. That's just the top of mind and I'm sure there are more. It looks like lots because you count each failure as confirming your bias, but if you correlate that against over 80% of small businesses failing then it's not a valid assumption unless you have data showing they fail at a higher rate. Don't get me wrong this is a shitty situation for sure, but it's always sad when an entrepreneur of any stripe fails. As an aside I think it's a more valuable conversation in general to look at the business models available to someone approaching this model of development and for us to make suggestions on how he could have built a successful product in this way and succeeded. My thoughts as a total outsider on this issue are going to be more general, but since it's my suggestion to do this I'll start. 1. I believe that when someone creates software in a broad usage category then consulting and especially training must be central to any business goals if you wish for similar projects to succeed. 2. Support contracts for large orgs are also very important because companies need security whether software is free or not. Without support contracts it's likely you won't get the kind of benefactors that are required to be financially sustainable. 3. Marketplaces seem to be in vogue as are extensions. These often work well for one very simple reason; people are paying either directly or indirectly (via their marketing of their extension) to promote your product/service. Nothing says win like others promoting your stuff for you. 4. Build a currated network of developers and promote them. Talk about them to others. Helping them is helping yourself. The more people or orgs investing in your marketplace developers the more stickiness a product will have. 5. Talk about the product everywhere. Conferences. Even start a conference around your vertical. Invite your dev circle to speak at the events. If nothing else it'll boost their careers. Anyway that's my off the back of the napkin 'just saying' for today. I hope it helps someone who's interested in following this gentleman to Foss joy and success. :)
  - pjmlp 8y ago
    The examples you cite work in enterprise space, not when selling desktop applications to regular consumers. Red-Hat explicitly moved away from desktop because of it. Even Ubuntu has given up and decided to pivot into IoT and server. Average steet consumer does not care about consulting, and if some kind of training is needed they ask around to friends or local teaching associations. The original author sees zero from those teachings and everyone in the business knows that desktop software books are more about building CV than actually living from them.
- pcwalton 8y ago
  It's hard to make money by selling desktop software, period. In 1995 it was a reasonable model, but there's a reason why everyone goes SaaS in 2018. The market isn't growing, and consumers aren't used to paying for typical desktop software anymore (with Mac a possible exception).
  - pjmlp 8y ago
    And SaaS is anything but open source most of the time, specially in the consumer space. We are back to X-Windows applications on dumb terminals, just prettier.