Follow

open protocols and open interfaces are more important than open source

corollary: free association and free interoperation are more important than free software

Show thread

@tindall I disagree with this one. Free software is about having total control and freedom over your own hardware. Make Noise Maths may be easy to integrate into a system, but you cannot change how it works if you want or need to.

@Kimimaru but you _could_ completely replace it in your system, with enough time and energy; nobody can stop you. in fact, it's ridiculously hard to find one for sale right now so i'm doing this.

whereas, for instance, you cannot completely replace Adobe products even with indefinite time and engineering resources because they own patents on components critical to interoperability

of course free and open to interoperation would be the ideal, but if i have to pick one i prefer interoperable.

@vandys I didn't say that open source _isn't_ important, just that open interfaces and protocols have a higher impact. I still run Linux, Firefox, LibreOffice, etc, and FOSS everything on my VPS.

@tindall
Definitely true (and needed!) in the creative computing world.

@tindall Sure. Let's implement a 100% compliant web browser **right now**.

@haitch Sarcasm is unhelpful. If you read the thread, you'll note that I call out exactly this kind of expansive and rapidly evolving API surface as a way to stifle innovation as well.

@tindall I've just read the thread.

I'm still unconvinced that proprietary software treated as a black box does anything to advance the ultimate goal of having a read/write system accessible by everyone.

By definition, it doesn't.

@haitch Rip out the engine from chrome or ff and go from there.

I like the webkit approach of having a webview only and let others implement the browsers around it. It would be great if webkit wouldn't suck.

On the pro side, qtwebengine seems to be exactly that and it works great in qutebrowser.

I think forcing a separation between webview and $else might be the most important part. Then evil company can build their spybrowser on it while a privacy respecting browser on the same engine would be possible.

Also we wouldn't have people crying for plugin apis that basically change everything.

@sh Webkit originated in the KDE project KHTML. It was FOSS before Apple started Webkit. (and in turn, Google expanded on it, adding V8 and a host of other things)

@haitch I know. And now the community can take blink back and refactor it into a serious competition webview with nice apis, no?

The web meanwhile is established enough that google would have a hard time with pushing incompatible changes that break the fork without pissing off a lot of their own customers.

well, a win either way...

@sh That doesn't fix the web, the web is still fundamentally broken if it allows for something like Facebook to be created.

@haitch This is a different problem domain. That boat has sailed.

Alternative protocols are popping up. Gopher pages declined until 3-4 years ago, now there are more gopherholes that ever. Gemini is on the horizon... things happen. The happen under the radar of big players, which is good.

In long term I see a commercial web based on facebook etc. and google where it feels like you walk through a mall.

But i also assume that the private web sector based on other protocols will grow and mature, which will generate a two class society. Except it's not money that seperates the classes but knowledge about how to operate this space and the clients.

To be honest, this could be a golden time with a net from people for people while corporations are still busy with "the other web".

@sh I couldn't disagree more.

I'm sorry, it would take me an endless stream of toots to explain why. You can find more in the historical discussion about these matters on this account going three years back or so.

We're not starting from the same frame of reference and there's an awful lot of information that you don't have available, it's simply not possible to discuss the complexity of this problem in lack all that information.

@haitch Do you have a blog that sums it up? I may do some scrolling on your account, but 3 years back is a lot.

@tindall I disagree, simply because any proprietary software can be changed (or taken away) by its copyright holder at any time, unilaterally, pulling the carpet out from under anyone who's implemented the open protocols or interfaces. That's a fundamental power imbalance. More on that: davelane.nz/proprietary

@tindall I think all those open elements are crucial. And I do agree that, as a starting point, open standards are more important that converting everything to #FOSS... although, I think that's the direction we need to be moving inexorably.

@lightweight @tindall I tend to agree more with the protocols/interfaces position than the source.

If we cannot communicate, we have nothing.

The tools we use for communication are less relevant, and can be varied, as long as they are intelligible.

The flip side however is that it's often Open Source that promotes protocols & interfaces in practice; however all too often these are de facto standards originating from proprietary software.

@yojimbo @tindall as I say, I think the power imbalance implicit in proprietary software is sceptic & should be removed from the computing, but I agree that the important first steps for moving to #FOSS-first is open standards/protocols/interfaces. (to that end, I wrote openstandards.nz)

@lightweight I was looking at the co-signers list at openstandards.nz/co-signers and noticed that there is a quadrupling effect to signs starting this year ( page 2; #107; 21 May 2020 ). you might want to look into that.

@tindall sounds like you think it is about choosing one over the other. We need both.

@tindall this note by @nightpool came across my timeline earlier : cybre.space/@nightpool/1046043.

I don't know if your note is related to that one, but both of you are saying the same thing. ( PS : I haven't read the paper linked to in that note yet; I've downloaded it for later though ).

And I agree too. To add to that argument, here's another paper that I've read in the past that is along the same lines : knightcolumbia.org/content/pro

Sign in to participate in the conversation
Cybrespace

cybrespace: the social hub of the information superhighway jack in to the mastodon fediverse today and surf the dataflow through our cybrepunk, slightly glitchy web portal support us on patreon or liberapay!