I think I'll be clear about this: I am not anti-Mozilla.

(Though putting Pocket in their addressbar rather than an RSS/Atom subscribe button does go against my design principles for encouraging a more decentralized web)

I think they are a positive influence, but they have too much influence to loose to be as radical as they pretend to be.

At the same time we do need a truly radical browser engine to help get us out of the sad software development we're in. I hope to be that!

I think I'll expand on what that design principle I have is: I do not integrate specific websites into my browser, whether it's my own or others. Always allow integrating multiple webservices into homepages, searchboxes, etc rather than just one.

I haven't managed to fully adhear to this yet, and I did make an exception to discourage reliance on YouTube.

But Mozilla doesn't know how else to make enough money to implement a "modern" browser engine. Neither do I.

Show thread

Might be answering a rhetorical question 

@alcinnz
You should've asked earlier :mario_jump:

The main problem is that browser design has always been a loss leader for web-based services. (1500 characters deleted) That doesn't mean you can't find a business model that will work, but having loss leaders in your target market means rethinking fundamentals about markets. In this case, for example, both for profit and nonprofit vectors are saturated, so you'd need a thoroughly postcapitalist approach

Distributing a browser that arrives on the end-user's desktop with a default configuration that approaches decisions they would make for themselves might be easier than attempting to eliminate any bias

But let me know if you're interested in a run down of the core ideas

Follow

@yaaps My approach so far is to tackle a poorly addressed niche, illustrating the benefits that have been discarded in the rush for webapps.

But I'm open to all ideas!

@alcinnz
You're an artisan. Occupying niches too small for economies of scale is how this precapitalist pattern survives and how it spawns new businesses under capitalism. It relies on a certain amount of privilege in that you need free time, skills, and personal/professional connections to succeed. in addition, the artisan is vulnerable to their own success. Replacing artisans with commodity labor is how capitalism grows

To scale up while maintaining autonomy, you need to establish a worker's collective. That provides a model to distribute income for skilled labor, as needed, without selling equity. Then you can crowd fund, if needed, while building a distributor network

Developing an independent distribution network is how you avoid becoming dependent on key donors. Your browser product will be customizable and unopinionated, with strong privacy defaults, which is not so friendly to most who would use it. Distributors will provide opinionated customizations to end users who share their opinions, without disabling privacy defaults or removing the ability to modify customizations. Your ideal distributors would be those providing community-owned alternatives to profit-driven content platforms - looking at internet access co-ops, user freedom CAPs, student government, clubs and hobby organizations. The general shape of the relationship looks something like the Linux distro marketplace, but maybe a different ratio of financial donations to code contributions going upstream. The idea, however, is to make it easy to repackage your browser with their settings and worthwhile to preserve your user-centric features and contribute

@yaaps I guess the obvious place to start, given the niche I've decided to fill, would be opensource hardware companies like Pine64. Or the distros they choose ship. Could be mutually beneficial.

Once I see if I can get beautiful auditory and TV remote experiences (by partially breaking compatibility), see if that can help them nudge into the phone/smartwatch/TV/etc marketplace.

Just trying to see how I can get from here to there...

@yaaps As for the issue of configurability, it's hard not to deliver that when building browser engines (as opposed browser chrome, like Odysseus). Though I doubt you'd get much from turning off the privacy features, given everything I could *never* afford to implement even if I wanted to.

Rhapsode's design in particular is turning out to be very elegant, and it'll be surprising how much milage you'd get from configuring it's default bookmarks...

@yaaps I will also say: This is a gigantic yak I chose to shave!

It's hard for me to think beyond Hephaestus, and before last week (when I studied Weasyprint, examined line counts, & had a seperate epiphany) that was seeming like a pipedream. I sure hope I get some help by then!

@alcinnz
You're doing really well, though. Just keep the razor sharp and that yak'll be bare in no time at all

@alcinnz
The original proposition was a "modern" browser (your quotes), so I was assuming that wouldn't be a 1 person job

Either way, EU and various national governments are offering grants to people developing alternatives to software in ecosystems controlled by US corporations and there are a growing number of crowd funding platforms

@yaaps I put those quote marks in to suggest that our concept of a modern browser is part of the problem. It's an unattainable goal for any group who hasn't already achieved it.

Yeah, I'm sure there'd be a government somewhere who'd be interesting in funding. I'll put out some feelers after I have Rhapsode roughly done. I'm surprised how fast that came together!

@alcinnz
Me, too. I thought I was going to have so much more time :yayclod:

@yaaps @alcinnz Or… we could… stop using a web browser for fucking everything.

When the neos have created a market from nothing, sometimes the solution requires leaving their market, rather than trying to play by the rules they have set.

@cy @yaaps @alcinnz

I'm clapping furiously here

native desktop apps are way, way superior !!

@cy Absolutely the stance I take!

It's achievable for me to build a web that work absolutely anywhere, it's by no means achievable nor worthwhile for me to build a web that does everything.

I need others to implement what should have long ago been discarded as out-of-scope for the web as native apps.

I'm attempting to change the rules!

@yaaps

@alcinnz @cy
I have strong opinions about what should be on the web - documents. And... my definition of document allows for interactive documents

Reasoning from lesser to greater, if I don't want to run arbitrary code in a web browser to read news, look at grocery sales papers, or report a broken dryer in the apartment laundry room, then I certainly don't want to install a native app for these things. Doubly so if "native" is actually rebranded Chrome with the web app baked in

So I think it's important to have a definition of document that includes forms, but to draw the line at executing arbitrary code or sending any information to the server that outside of that specifically requested of the user. In order to accomplish that, there would need to be a browser engine that is not designed by those working for publishers

As a bonus, such a browser engine might actually be useful to render HTML documents within a native application without the bloat of an engine designed to cope with the web as an execution environment. Thinking ActivityPub client and game clients, here

@yaaps @alcinnz @cy

Right now it looks like there are only two options to, say, buy things from an online store:

1. Use an interactive website of that store
2. Use a native app of that store

And option 2 is more intrusive than option 1. But there's a largely overlooked third option:

3. Have a common protocol for all online stores, and use a native online shopping client of your choice.

@yaaps @alcinnz @cy

Now, option 3 is hard because it requires competing parties to agree on a common protocol.

It is also unattractive to the stores, because it's not a branded experience - the shops cannot influence your UX.

It'd be good for users, but I have no idea how to push self-interested shops to adopt it.

Of course same goes for any other service - chat, ticket booking, etc.

@wolf480pl @yaaps @cy In the case of online shopping, I can't see how my work excludes option 1.

Sure I'll break many existing ecommerce sites, but I'm planning to allow cookies to be set (only) in response to form submissions. So an "add to cart" button can still work.

Love to have builtin form input for payments...

@alcinnz @yaaps @cy
I'm not saying it'll make things impossible in this particular usecase. Forms are pretty great!

But I'm pretty sure there are some usecases which can't have option 1 without arbitrary code execution and XHR.

And in any case, I'd love to see both groups of usecases to eventually move to a standardized protocol.

@alcinnz @wolf480pl @yaaps Well obviously you should expect resistence from every single dishonest ecommerce merchant, but they shouldn’t have enough power and influence to reject your system and all the people who support it, unless there was some kind of wealth disparity or something.

Have a look at IATA's NDC (New Distribution Capability) whose intent is just that - ability to search available "products" (tickets) and sell tickets on 3rd party via airlines NDC API iata.org/en/programs/airline-d @wolf480pl @yaaps @alcinnz @cy

Sign in to participate in the conversation
FLOSS.social

For people who care about, support, or build Free, Libre, and Open Source Software (FLOSS).