[archived] On GPL without contributor licenses, the App Store, and future of Pulse

Development
21

I was personally glad to see the move to GPL as for me that’s the sign of a project that will last, often beyond changes of stewardship. In contrast to some who write-off GPL software, I will write-off non-GPL software when deciding what to invest time / energy / and code in. I say this just to show that not everyone is against the GPL. (Though of course I respect everyone’s views)

With regards the App store, I am happy to see Apple unable to benefit from the wealth of GPL software as it is a small encouragement for end users to shun their products. The solution to the App store problem in my view is for those who wish to see syncthing supported on such devices to create their own implementation of the protocol and put it under a license compatible with Apple’s policies. Not ideal from ind.ie’s point of view, but a workable solution.

I am grateful to all involved in syncthing for such a great contribution to our world. And equally I have great faith in Aral’s vision for a more equitable future for all regardless of technical ability and wish him all the best in proving the model works.

I hope you don’t mind me sharing my views on the project with you all, I know time is precious! Oh and… I’m celebrating my first minor contribution to the codebase going out in a release :smile: Yay!

6 Likes
22

Shots fired! :smile:

23

FWIW, the protocol implementation is now separate and MIT licensed, so at least that part is free for anyone to look at, copy, modify, print, burn, yell at and so on. :slight_smile:

https://github.com/syncthing/protocol

Hopefully some cleanup and docs to follow to make it more of a genuinely reusable component, but in the meantime it’s the code to go with the specs.

3 Likes
24

FWIW, if a project grows, it might make sense to introduce a contributor license agreement (CLA) to be able to change the license at a later point in time. The GPL might not be the right license forever. There could e.g. be a legal bug in the version of the GPL used. An agreement with a nonexclusive license could be sufficient for the purpose of changing the license later.

But in any case the decision to introduce a CLA should be carefully planed, as it involves a lot of difficult decisions and requires a certain infrastructure. The most important decisions would probably be: Who decides a license change and how? and Who do the contributors license their rights to? (Ideally a non-profit entity.) Before these questions can be agreed on it probably makes little sense to debate much about whether a CLA makes sense. But at some point it will be too late. Because the introduction of a CLA, just like any change of licnese without a CLA, ideally requires everyone’s consent. The VLC experience when trying to relicense is a good example for the difficulties in getting this when a project has grown.

1 Like
25

Just to update you all, we have gone back and reforked from the MIT-licensed version. We’re building upon and using that version in Heartbeat.

We’re also no longer supporting Pulse as a separate product, although anyone can still download and use it. We’ve suggested that people look at Syncthing instead if they want a standalone file synchronisation app.

You can read the full update here.

Best of luck with everything and I’m sorry it didn’t work out.

26

I know that my answer is completely off-topic here, and I apologize in advance. However, since you posted a link to your article and there is no way to comment, I’ll do it (shortly) here.

The paragraph “Why Mac?” is a complete non-sense. Apple is a spying company. For the record, they are tracking their iPhone/iPad users, and are also spying the searches in Yosemite. You compare Apple to Google, but you know the comparison with Microsoft or Linux would not keep up very long. The only honest sentence I read is the following:

Finally, all of us at Ind.ie use Macs and so we are both familiar with the culture of the platform and we will be eating our own dog food, as it were.

You are a very small team and all Mac users, so you develop for Mac. This pragmatism justifies your choice.

Again, sorry for the off-topic.

1 Like
27

Apple is a closed/proprietary company and thus not ideal in the long-term (and thus why our goal is to eventually build our own platform). However, you must understand that the question isn’t “can they spy on you if they want to?” (sure, yes) but rather “is it in their interests to spy on you?"

For Google, yes. Spying on you is how they primarily make their money.

For Apple, no. Selling you products is how they make their money.

Now think about what’s happening in the world. More and more people are waking up to corporate surveillance. This includes policymakers and politicians. The corporations are countering by increasing their lobbying (“ok, you see us for what we are now, here’s some money”). Regardless, there will be an increasing amount of regulation. Some of that regulation could seriously harm Google (which is why they’re spending so much on lobbying). When I talked to Eric Schmidt last year, he said that regulation “could kill Google”. Not immediately, but slowly, over time, could kill what it can become.

So take a world where privacy starts becoming competitive advantage. And yet it is the one area that Google cannot compete in. (Not won’t but can’t, as it is mutually exclusive with their business model.) They can only compete in the illusion of privacy and they can compete in security, which is a related but very different thing. What they cannot compete on is actual privacy.

If Google stops spying on people, they will go bankrupt. (And if they even thought about altering their core business model, they would be sued by their shareholders.)

Apple, on the other hand, can compete on privacy because their core revenue stream is not tied to monetising data.

So answer me this: if you have a competitive advantage that your main competitor(s) cannot compete on, would you throw it away?

Only if you’re an idiot.

And something tells me Tim Cook isn’t an idiot.

1 Like
28

I also want to comment on your Article. Why are there no comments? (and please without disqus)

29

I LOL at the whole “Apple isn’t a spyware company” because that’s just being naive.

2 Likes
30

Is the code you’re writing for Heartbeat not as easily portable to other platforms as you had first thought?

The concept of Heartbeat is great, I would really use a decentralised Facebook-like platform, especially if it had options for event handling, contact syncing and synchronous communication, I would have liked to have seen it available to more platforms early on, but I suppose that if this is going to be open sourced in the future, someone in the community might be able to do that.

31

(Bringing this slightly back on topic…)

That’s one way. :) As I remember we talked about at the start, I think it’s more likely you should remove code rather than add it, to work well for your underlying requirements. Will you be publishing source for the result?

(Edit: I previously asked whether the existing Pulse development was going to be stopped, which has been answered in practice by the repo being renamed to ...-legacy. I see there’s also activity on the Swift port, which is neat!)

32

Wow.

I had high hopes for ind.ie.

Carry on, Mr. Bork. You’re doing good work.

You have my sympathies for putting up with such abuses.

1 Like
33

Guys, lets keep this thread about the original topic (as far as that’s still relevant, which is not very…) and not about the latest Ind.ie developments. Specifically abuse of any kind should not be a thing here, and both me and Aral may have gotten a bit excited above but there’s no reason to pile onto that.

7 Likes
34

The folliowing reply is a bit old - about last christmas:

That’s a discussion!

Now that I read through it, I’ll try to contribute something productive.

  • I see Syncthing as a synchronization tool. It replaced Ubuntu One for me.
  • I like the idea of a decentralized social network. (Hey! I’m on Diaspora*!) But I don’t think this should be done by synchronizing files to each and every machine - build many servers run by many people and make them interconnect.
  • I didn’t understand the post as an insult but more regarding open source software in general. But after reading a LinkedIn and a Wikipedia page, I see how this can be interpreted as a personal insult. (How and if people are responsible for things other people do with their inventions, is an interesting question, for sure. But I don’t think we should or could answer it here.) I can’t decide, whether this was an insult or not - but targeting someone personally doesn’t help the discussion - for sure.

Now today:

Well, I’m glad that Syncthing is licensed under the GPL - and I clearly can’t see why this could be a problem. The only half-a-problem I can imagine is someone wanting to make changes, not distribute the source and sell the result. I’m no expert for Apple, but does the GPL really prevent you from publishing an app there? (That would be a reason not publish apps there. Cydia, anyone?) I’m not against selling software. But a trustworthy social network has to be open-source. Otherwise, it just isn’t trustworthy anymore. (And this doesn’t collide with selling. You are able to sell open-source software!)

35

This topic is no longer relevant.