Android or iPhone — either Google or Apple delivers our messages — surely? You don't accept that?
Time I Learned: there are freedom-respecting phones.
People who do not want to depend on Google or have them control our devices are using android-compatible but not google-controlled phones, a.k.a. “degoogled phones”. We have been asking (ourselves) for several years if we can have google-free push notifications. Thanks to the developers of the UnifiedPush standard, the answer is now, “yes!”
UnifiedPush open-standard push messaging complements degoogled android-compatible phone OS's such as LineageOS.
People who do not want to depend on Google or have them control our devices are using android-compatible but not google-controlled phones, a.k.a. “degoogled phones”. We have been asking (ourselves) for several years if we can have google-free push notifications. Thanks to the developers of the UnifiedPush standard, the answer is now, “yes!”
The open standard UnifiedPush.org has now been created. While not a large number yet, a useful handful of apps already support UnifiedPush, including several matrix and fediverse apps. For its servers and the associated client-side “distributor” component, there are multiple successful implementations deployed.
More and more I find myself using the fediverse (based on ActivityPub) alongside Matrix, and the two protocols complement each other nicely in some ways. In other ways their differences seem needless and jarring, only getting in the way. I'd like to be able to offer my contacts the free choice of which protocol to use, depending on what kind of communication they intend (public-ish versus private or group-ish), without everything else being different — especially the user identifier/addressing but also many lesser issues such as the existence of user profiles.
I have long remained hopeful we'd get use profiles eventually. Conceptually it's not a big deal. So where are we now and where are we going?
This article opens with the well known idea of using (free/libre/open) tech for teaching (free/libre/open) tech skills, but it is not about that, at least not directly.
Tandoor is a self-hosted open source recipe manager. I have been running it in my YunoHost test server, in order to try it out, for about a year, collecting 21 recipes in it.
Basically I'm not satisfied with it, for my needs.
It's quite nice in some ways but terribly limited to one particular fussy attempt to organise a recipe in a particular way with steps, ingredients categorised by name and quantity, etc. There's no way to include recipes that don't fit this format. No way to include a web page (that doesn't import neatly into this very precise format) or a text document that I wrote or a PDF scan of a paper page. No way even to represent “fuzzy” ingredients like “some herbs of your choice” or “about 2 or 3 cloves of garlic”. Each ingredient has to have a number and a unit specified, and there are only hacky work-arounds like defining the word “some” as a custom unit. So I'm still running it for fun, for a small handful of recipes I imported (sometimes with awkward manual clean-up required) from web sites, while most of my recipes are still on scraps of paper or PDF scans or books or web sites that don't import.
Open tech, be afraid. Be very afraid. Microsoft owns both Visual Studio Code “VSCode” and MS-GitHub, two intertwined and utterly proprietary product-service ecosystems with a bit of open-source in their core to lure us in. Because they love open source? Yeah, no.
Soon after leaving GitPod whose technology links the two, Geoffrey Hunt last year explained their strategy and what it's doing to our open tech world, in a great and “harrowing” article, “Visual Studio Code is designed to fracture”
https://ghuntley.com/fracture/
“The future of software development tooling that is being built is closed as ****, and people seem to be okay with it...”
This is why MS-GitHub is not our friend.
This is why falling for their trick, disguising MS-VSCode as a neat “free” editor, will come back and haunt and hurt us.
Vendor lock-in double-whammy. Using open source as “a financial weapon”.
”... the biggest challenge for Gitpod, GitLab, Datacoves, OpenBB, Foam, et al lies ahead – developing open language tooling for each community where Microsoft has forked the communities over to proprietary language servers...”
If we have a grain of public spirit, if we are motivated at all by the Freedom that's supposed to be afforded by Free-Libre Open-Source Software, we must #GiveUpGithub, we must recognise the trap, we must choose truly open #FreedomTech.
We tend to think of Google Search as the gold standard, the comprehensive, personalised, convenient, quick and reliable option. The one for getting things done. The experience that other search engines can only aspire to.
But, as we know, Google Search is designed around the financial goals of the advertising business. Can we understand just how far that misaligned incentive has warped the whole experience? What if a search experience were designed in a different way, around what's good for us, what's important to us, our real values? I don't mean just the same kind of search experience but with adverts stripped out. I mean if the whole system, from content publishing through to browsers and apps, were redesigned. How unimaginably different might that look? And as we obviously can't jump straight to that world, what insights does this give us about improvements we could seek in our current world?
Anyone still thinking Google Search is “good”, after learning about what is going on behind the scenes, is missing a perspective on what “good for us” would really look like.
“Have you ever wondered why every cooking recipe on the web has a twenty page biopic preamble? Because Google likes it better that way.”
Announced by @davx5app@fosstodon.org, the good folks at DAVx5.com who make the libre/open sync for CalDAV/CardDav/WebDAV, that connects Android/Outlook/Thunderbird to standard calendars, address books and file shares: they are now looking at designing a “push notification” standard for the DAV family of libre (open) standards.
Yay! Fantastic!
With push support, my family would get instant, efficient updates to our shared calendar and address book whenever any of us add an entry, for example.
The most interesting thing about this, for those of us who care about liberty and choice, is that the push delivery system should not be locked in to google/apple but should be able to use the UnifiedPush.org open standard that lets each end-user choose their preferred push provider. (See my other posts on UnifiedPush.)
I was just preparing a Merge Request to contribute upstream, when I noted,
You can review my merge request in the web UI at my TraxLab (gitlab) repo. Obviously you can't click the “Merge” button (until Forge Federation is done — there's an awesome project to check out).
It still grieves me that open source devs push me into working with Microsoft Github. Sure I understand the argument to use it “because it's convenient right now and 'everyone' is there” but to me there's a more important value I wish to uphold:
Millions of Free Software developers forgot why it matters to own their tools.
Last week I was setting up Draupnir on my matrix test rig, in order to become familiar with Draupnir deployment before I integrate it with PubHubs.
Now I need to address end-to-end-encryption (E2EE). PubHubs exclusively uses encrypted matrix rooms, and Draupnir doesn't yet have E2EE functionality built-in. (Why is that? Moderation in public rooms is Draupnir's main use case, and for several reasons public matrix rooms are usually not encrypted. However PubHubs is different.)
There is a generic solution for adding E2EE to a matrix bot, and it's called Pantalaimon, an “E2EE aware proxy daemon for matrix clients.” So this week I'm setting up Pantalaimon.
