[HN Gopher] Releasing my tools under the MIT License was probabl...
___________________________________________________________________
Releasing my tools under the MIT License was probably a mistake
(2023)
Author : marbu
Score : 69 points
Date : 2024-02-17 21:06 UTC (1 hours ago)
(HTM) web link (donatstudios.com)
(TXT) w3m dump (donatstudios.com)
| rad_gruchalski wrote:
| I can relate. After some years I first switched to the Apache 2
| license, later for some code AGPLv3. My stuff is mostly garbage
| and some software debt but who knows ;)
|
| Now I start leaning towards: if one doesn't want others to use
| it, don't release it.
| LadyCailin wrote:
| > they don't credit me as the author or provide any sort of link
| back
|
| But this is simply not true. MIT requires keeping the copyright
| notice intact, which would be a credit. People that aren't going
| to follow this requirement weren't going to follow the GPL or
| whatever alternative you pick either, so either sue them or don't
| worry about which one you picked exactly.
| jonathankoren wrote:
| Agreed. His problem isn't the license. His problem is thinking
| a text file is going to stop a bad actor.
| billti wrote:
| He covers this in a couple places in the post. There is no
| requirement under MIT to redistribute the source for any
| changes you make or anything you build with it.
|
| If they redistribute the source then yes, but that's not the
| concern in the post.
| csande17 wrote:
| Including a comment in the JavaScript file is broadly
| considered good enough to satisfy this requirement of the MIT
| license, even though most people won't ever see it.
|
| GPLv3, on the other hand, is much more explicit about saying
| that the copyright notice must appear in the actual user
| interface of the application.
| tomcam wrote:
| tldr; OP felt that releasing libraries under MIT License
| benefited the community, but releasing apps was a mistake because
| other sites bested them in the SEO game. That probably caused a
| Bing blackout, and certainly meant losing in SEO to crapware-
| filled sites.
|
| I'm thinking the optimal course would be a GPL release +
| trademarking the software name so that there could be more
| control about attribution and what sites get to use the name?
| justinjlynn wrote:
| Yes. The (A)GPL is there for a good reason (in part, this one -
| ensuring one's work and other's work on it remains free and
| open source and commercial freeloaders can't get a free ride),
| and trademark law ensures you retain control of your software's
| brand. MIT and BSD... well, look where they come from - they're
| not designed with those purposes in mind. If you care about an
| aspect of a licensing solution, use a license designed and fit
| for purpose - just as you'd use a library designed and fit for
| purpose.
| AshamedCaptain wrote:
| > Most irksome of all, in a fair number of cases they sit
| centrally on pages covered in ads and SEO keywords. My tools are
| being associated with a genuinely bad user experience.
|
| For the record, any license that does not allow users to do that
| would NOT be a free software license.
| dflock wrote:
| Scammers gonna scam, whatever license you use - but, that said,
| this is the exact use case that the AGPL license is made for.
| pcthrowaway wrote:
| > Many of them have made minor or major modifications to the
| tools, and next to none provide the source to those
| modifications.
|
| > I am considering relicensing my tools under some sort of
| Attribution-ShareAlike license similar to the BY-SA the content
| on this site is licensed under.
|
| Wouldn't the LGPL be well-suited to this?
| eastbound wrote:
| I think you mean the Affero license:
| https://en.m.wikipedia.org/wiki/Affero_General_Public_Licens...
| heavyset_go wrote:
| Yes, and in some cases the GPL and AGPL.
| freedomben wrote:
| > _I want you to use things I 've written. On top of that I
| don't believe it's my place to force you to then open source
| things you have written that expand upon my source code._
|
| I'm a big proponent of the GPL and the AGPL, but no they
| don't sound like good solutions to the author's problems. It
| might solve the attribution issue but it's going to go
| counter to the author's other goals.
| wmf wrote:
| The attribution required by GPL/LGPL is fairly weak and in
| practice doesn't prevent the kind of "exploitation" discussed
| here. Ultimately if you don't market your work you _will_ get
| steamrolled by people who do; I 'm not sure this can or should
| be fixed.
| OJFord wrote:
| I'm not going to go looking for them, but the impression I get of
| the sort of copycatters described is that they really won't care
| what the licence is, if the source is available they'll be there
| anyway - the blog post will just be complaining that it's against
| the terms of the licence (and probably not pursuing legal action)
| instead.
|
| I don't think it really matters. These things will exist, anyone
| who matters will realise they're not legit. They won't make
| significant sales (without significant added value) it won't
| detract from your reputation; etc.
| csande17 wrote:
| If they were actually violating the license, it'd be pretty
| easy and cheap to send DMCA takedowns to search engines,
| wouldn't it? (And possibly also whoever's hosting the copycat
| pages.)
| znpy wrote:
| I get your point, but i guess it's sadder to see people
| complying with the license terms in an assholish manner than
| see people completely breaching the terms of the license.
| OJFord wrote:
| But I suppose I'm saying I don't think they _have_ that
| arseholish manner - they are simply replicating it because it
| is available.
|
| I.e. if you don't like it the solution is not AGPL or source
| available but no reuse allowrd or whatever, it's closed
| source.
| dang wrote:
| Discussed (a bit) at the time:
|
| _Releasing my tools under the MIT License was probably a
| mistake_ - https://news.ycombinator.com/item?id=37111145 - Aug
| 2023 (7 comments)
| albertzeyer wrote:
| So, what does the author actually wants from the licence?
|
| Is it ok that other people take the code, modify it but don't
| open the modifications? If not, then GPL or AGPL. If you want
| that they can still build sth around it, but otherwise not
| modifying your library, then LGPL.
|
| Or is this ok, but the main issue is no attribution to the
| original source? BSD licence maybe? Or Apache? Or what else?
| hiAndrewQuinn wrote:
| I basically have two modes for releasing code: All rights
| reserved, or public domain / CC0. My reasons are pragmatic. The
| latter improves the former by letting future me shamelessly
| plagiarize past me with zero responsibilities to point it out to
| anyone.
| Scubabear68 wrote:
| The specific problem the author mentions likely would not be
| solved by a more restrictive license. SEO squatters take whatever
| they want, and I doubt it would be worth it to sue them for
| breaking license terms.
| reactordev wrote:
| First, you wanted people to use your tools and you gave them away
| for free under the MIT license.
|
| Now you're complaining that people are using your tools.
|
| You can't have your cake and eat it too.
| donatj wrote:
| Author here, and I agree. In the post I'm just lamenting my
| past choices. I gave people leeway and they used it. That is my
| failing, not theirs.
| RobotToaster wrote:
| >I am considering relicensing my tools under some sort of
| Attribution-ShareAlike license similar to the BY-SA the content
| on this site is licensed under.
|
| Please don't use CC licenses for code, it's not what they are
| designed for and the CC actively discourages it[0]. Consider
| using the AGPL[1] or similar instead.
|
| [0] https://creativecommons.org/faq/#can-i-apply-a-creative-
| comm...
|
| [1] https://www.gnu.org/licenses/why-affero-gpl.html
| znpy wrote:
| Not sure they can change the license at all?
|
| The author doesn't own copyright for the code changes they
| accepted over the years.
|
| But kudos to the author for acknowledging they picked up a dumb
| license. Sad not to see the GPL or AGPL considered though.
| an1sotropy wrote:
| The author is sharing second thoughts about using the MIT license
| and yes, bad actors are going to break bad, but the point of
| licensing is to control re-use within the (enforceable) legal
| framework of copyright. Reciprocal licenses (thanks Lawrence
| Rosen[1] for that term less charged than copyleft or viral) cede
| less control, and provide more footholds for enforcement.
| Remember that GPL has (sometimes) worked as intended in
| adversarial commericial settings [2,3].
|
| [1] https://www.oreilly.com/library/view/open-source-
| licensing/0...
|
| [2] https://www.tp-link.com/us/support/gpl-code/
|
| [3] https://www.zdnet.com/article/software-freedom-
| conservancy-w...
| donatj wrote:
| Author here. This was just a frustrated rant because these people
| serving my circle generator and my .htaccess rewrite generator on
| sites plastered with ads get more traffic than I do. I'd take it
| as with a grain of sand.
|
| I frankly used to to have a little cottage industry that helped
| me pay the bills from people finding my rewrite generator, not
| knowing what they're doing, and reaching out for help with their
| htaccess. It's been a couple years now since anyone has reached
| out. Part of that decline is clearly Apache becoming less
| relevant. The other part is that I've fallen way down the SEO
| ranks, frustratingly behind people hosting my own tool.
|
| Everything is still MIT and by all likelihood going to stay that
| way.
| adrr wrote:
| If they are modifying the frontend so it's their own code, no
| copy left license would work because no copyright would be
| triggered. Output of the tool wouldn't be covered.
___________________________________________________________________
(page generated 2024-02-17 23:00 UTC)