Stream - article - Jacky Alciné

An Idea on Following People (or Organizations or Whomever) in the IndieWeb for Lwa

After a bit of discussion in the IndieWeb channels (start here and end about here), I think I have an idea on how I want Lwa to handle "people". Tantek made a good point about optimizing the experience to follow people (or any sort of "entity" that isn't specifically a site) - that's a behavior people are familiar with and it's natural. This got me thinking about how I want this to look in my social reader.

Setting up the Environment

This requires a bit of contextual setup - where would someone find someone to "follow"? I'll optimize this to be in the case of someone who's already logged into Lwa (which assumes that they have a IndieWeb site) and are already following some people.

A preview of a post in Lwa.

From the screenshot above, I can think of the following ways to follow someone from this screenshot of a post in the feed for Lwa:

Adding a button next to the post-type icon to follow the person.
Adding a pop-over modal that'd provide a "best-effort" representation of the entity that made that post.
Providing an intent to follow page that'll provide information on that user

The first two can be collapsed into a on-page experience that'll present some information about the content's author via a modal of some sorts. Ideally, that'll have information about them (name, photo and whatever else is semantically relevant). There, we can have a page that'll heads to the "intent to follow" page mentioned in the last step. This is where things get a bit interesting.

An Intent to Follow

The intent to follow page will be vital. I haven't built one yet but it'll be something that'll combine the following sources of data:

a "best effort" representation of the author
conditionally presenting the post that brought us here
a way to view the multiple feeds presented by the author

I'm thinking about making this page accessible by either providing the author URL itself or the contextual post to show. The latter will be something that helps people just finding content. The method of composing the author representation will be something of a collection of the following with (unspecified) weights:

fetching rel=me information from the provided URL
fetching the representative h-card of the URL
opting to use Microformats over anything else when it's available (specifically, Microformats, Activity Streams and then whatever is left over)

Building a list of the feeds for this user would be done the same way. One thing that'll help with optimization of these feeds is, in the event that they have a Microformats2 site, we can collapse syndicated posts into the site's post so the feed doesn't look too noisy. With these feeds collected, we can do some light metrics crunching about post frequency, the average interaction rate (if any) for posts in a feed. With that, we can give the user in question a choice of what feed that they'd like to follow (and into which channel). I don't have an immediate mock-up of that page but I can see it being a mix of Twitter's user profile pages and the now-defunct Twitter intent-to-follow pages. There's a case to be made to allow for a "catch-all" channel for people to use but I think I'll make that an opt-in function of Lwa; having such a thing tends to enable more lock-in.

I'll make another post once I get around to the demo of this feature in Lwa. For now, I'm down to get feedback about such a flow!

2020-06-09T13:05:30.97684-07:00

Tags:
lwa

On Owning Data Apps Use and the IndieWeb

So I came across this post about owning data from apps we use a lot. As I'm reading it, my head's bopping up and down with agreement - I strongly believe that data sovereignty is a important human right; especially with large companies actively using it to dismantle facets of society. In the post, there's a lot of things that companies like Google and Apple are inching towards implementing in their data warehouses to learn even more about you - something I am not an advocate of as someone who interfaces with activists who have to defend themselves against state actors and people who work at companies that are doing this vacuuming. However, the approach that the poster takes requires a bit of programming that might be not friendly for most people.

Now, I don't mind writing code to get off the ground but I don't want to keep writing code to do things. I'd like to have some sort of plug-and-play approach to this and that's where I think the IndieWeb comes into play. Though still optimizing for personal use, the IndieWeb has a lot of people working on ways to track things around what they read, where they go and what they interact with on the Web. I know that I'd love to capture more information so I don't have to rely on companies to keep them up or eventually sell them to Google or Apple (FitBit and Dark Sky). This isn't easy though because of these platforms restricting the level of control people have over the devices in the name of "security".

The original poster mentioned something that I actually use to this day to handle a lot of this stuff: OAuth (namely OAuth2). In the IndieWeb, we lean on an extension (with loosely scoping) of OAuth called IndieAuth. It gives you all of the security and reliability of OAuth2 while reversing the authority to your own choosing. We also use things like Micropub and Microsub to handle how we do storing of content (namely contacts, tags and posts) and what we subscribe to as well as what we follow. I use both of those daily to handle how I interact with the Web. The thing is, at least from the poster's perspective, most of this stuff requires way too much wiring and coding for people to even use. And I agree. There's no pressing interest in the community to get people who don't have a high comfort in writing code since this is not only a community-led effort but one without any big funding behind it. I think that initially can be a good thing because it prevents too much pollution by direction of funds but that's a different conversation.

That post definitely felt like a form of validation that there's interest in this and that the stuff I'm working on can nudge people towards it. I want this to be as "easy" as it was to hop on sites like Tumblr or AIM and communicate with people you know or find new ones. Obviously, that ease of use came with years of work and research but most of that has been done and the hardest part now is the design and implementation - working on user-centric, people-first systems tends to remove the whole "scale scale scale" problem. I'm starting this with a "plumbing level" service and I'll work my way up to meet people with Lighthouse. What do you want to see?

2020-05-24T00:17:10.86664-07:00 • 4 interactions

Tags:
indieweb
thoughts

Backing Up My Life and Sending My Machine Away

I'm in the process of shipping my personal laptop to Dell for repairs. At first, I was a bit worried that I'd be out of a machine for work (I use it for client work at black.af as well) but I have a PC that I've dedicated largely for media consumption (gaming and video storage). That made me wonder how much of my actual online life do I have backed up in the event that my main laptop was out of commission. I took some steps below to move my "vital" infrastructure over to a closer location.

I copied my GPG keyring into my Yubikey. These keys are useful only really to me with one publicly known key and another key I keep for personal encryption that isn't shared anywhere.
I moved any files I wanted to keep long term into my personal install of Nextcloud. I've been using Nextcloud since 2016, ownCloud since 2015 so everything I need and want is stored in there. That server also has reproducible backups (test your backups monthly!)
I started uninstalling a few GUI applications that I used a lot and checking if I could back up their settings. This ended up being my terminal, IRC client, e-mail, calendar, code trees resulting in gigabytes of information having to been sent over rsync. A good time to raise a ticket with KDE about making the serialization of one's system settings an actual practice - this way, I could keep some settings in sync with others across devices. This is going to become more prevalent as Plasma Mobile is likely my mobile operating system of choice in 2020.

The most complicated part was probably copying my files over that I didn't keep synchronized under Nextcloud's watch. This was mostly the code of projects I work on which produced and all of the assets in cache as a result of it (curse you Node modules and projects that can't support the link flag).

My primary laptop for two years

I don't know why but I got slightly emotional packing up my laptop and putting it in the box, lol. I'm eager to have it up and running soon. Safe travels!

2019-12-04T16:53:07.69135-08:00 • 2 interactions

Tags:
thoughts

Ambitions for 2020: A IndieWeb Look

With the new year looming around the corner, I want to help push things that encourage a more people centered Web to arise. We've returned to the land of the mainframes and it's evidently not something that we need to keep around if we're invested in people. The following list is a short one of things I'd like to have for the IndieWeb to help move the needle:

Contribute more to the Android client for Indigineous
Experiment with my own reader for Android
Publish the mobile client for Fortress
Work on pushing streaming support for Microsub
Re-attempt launching IWC Oakland
Completely deactivate all of my Facebook accounts
Build better interoperability between ActivityPub and IndieWeb communities
Explore bridging the worlds of Secure Scuttlebutt and the IndieWeb

There's definitely a host of more things I want to get done in 2020 but if/when I get these done, I'll be more than glad. Here's to a new year around people!

2019-12-01T00:22:31.49862-08:00

My IndieWeb Challenge!

I'm going to be participating in the IndieWeb challenge in December this year. Most of my work is going to focus around building up services that I'd like to use for the IndieWeb and can help use to encourage use of it. This means I'll be doubling down on building Lwa and Fortress to bring a polished experience for browsing social feeds and authorization. If I also find time, I might look into implementing collection posts for Koype so I can finally port over my FAQ! That coupled with the ability to have custom pages in Koype would be the last additions to getting my experience to be completely IndieWeb.

I also do plan to do some Wiki gardening for older pages to help keep things relevant for those pursuing. Here's to the IndieWeb Challenge!

2019-11-30T23:56:07.67776-08:00 • 2 interactions

Tags:
indieweb

Working for an Ethical Software Company

Ethics. A governing set of rules that help people determine conduct, according to Wikipedia. These kind of rules have become more and more important as software continues to "eat the world". Recently, Paul Jarvis asked an open(?) question about what it means to be an ethical software company. Which is always going to be subjective - some people consider some things to be "right" or "wrong". For example, imprisonment of any being is wrong. Angela Davis taught me that. But we have companies who have no problem working on a mission to accelerate and "effectively" put people into prisons or detention centers. Ethics - subjective as hell.

For any company that wants to come off as an ethical one and walk the walk of it, they have to understand that their impact doesn't end on deploys. It continues into the action that they take with the money (and resources) they collect from users to continue their livelihood. This line is one that tech (for some reason) is willing to ignore in some cases and what makes the actual difference between an ethical company for show and one in the paint.

It has to start with the people in the company. An ethical company values all of the work produced by its workers (think cooperative economics). This insures that by design, the company is stabilizing for the well-being of its workers and not for vacuuming excess profits for shareholders. This focus on sustainable growth spills over into other things like:

collecting and communicating data from users when it's necessary, not because you can
listening and responding to feedback from users about how it impacts their lives due to having a more direct connection to the success of the company
responding to events and supporting the well-being of workers when it comes to their lives

This definitely comes off as "this company is creeping into their personal life" and yeah, that's a bit of the idea. When companies decided to become "the new social interaction layer" or "the new movie theater", they took on the responsibilities that come with these deep social vectors of people's lives. That obligation has been taken extremely lightly in the tech sector and has resulted in a lot of mistakes and harm for the sake of "innovation" and "scale".

Companies won't be perfect - nothing really is. But they can do better (we're past the saying part). And this ties into what Paul was really asking - what do they do?

They stand behind the values that they profess. If you're a company that works to connect people; don't fund or back things that actively disconnect people from their families. If you're aiming to be a transport layer of ideas, understanding that one user is not representative of all is necessary and creating tooling for such a landscape is necessary. You truly have to put yourself in the lives of others (actual empathy) and understand how supporting and working with organizations at scale can impact your users. It's doubly ironic when companies claim to want to reach into "developing worlds" and are indirectly making their lives more difficult. This is an obvious jab at companies like Microsoft, Amazon, Google, Facebook and whatever letter you can fit in the FAMG acronym.

Donating the money away doesn't matter if it's not proportional to what you take in nor does it undo the harm initially caused. It's like a conditional apology (those aren't apologies).

For those who don't wanna read the wall of text above, just try to apply the following:

Give users complete say over their data on your platforms
Invest time into not being a member of the Attention Economy
Understand what you spend money and/or invest in
Be as transparent with your choices to your users as you would be with stakeholders (users are the validation of your product after all)
Pay your workers equitably, fairly and transparently.
Avoid surveillance capitalism (it's anti-ethical by design - unless your ethics have no problem with abuse of individual privacy).
Grow from the mistakes and listen to people who are negatively impacted by those mistakes.

2019-11-07T12:49:18.70415-08:00 • 11 interactions

Tags:
thoughts

The Future of the Intersection of PoC and F/LOSS

Kat proposed a question that gave me a bit of pause (thanks Josh for highlighting it). I figured for a question with this kind of weight deserved a longer space.

The F/LOSS space gave me my stay into technology. I started on Windows ME because that's what came on the Compaq my mom bought. But when I couldn't update the system past Vista and it begun to really slow down, Ubuntu was there - a bastion project that shows what Open Source software can do when we have people and companies backing it. I have a lot of faith in the power of free and open source software. But there's a reckoning that needs to occur.

Free software is defined by the Free Software Foundation as "software that gives you the user the freedom to share, study and modify it". It's commonly described as "free as in speech". Open source software's a bit more defined as code that's licensed as freely available, permitting anti-discriminatory distribution, providing non-encumbering and non-specific application to the software being used and technology neutral. With these definitions, you'd think that organizations like these and their backers would be very open and willing to take on people who are conventionally pushed out of places that gate-keep knowledge and resources.

I rarely had problems online until people saw the face associated with it. I've gotten unlogged comments on FreeNode about people being "shocked" that I'm Black. On the flip side, I felt safer and more comfortable seeing and working with very welcoming people (hey Paul!) that kept me in the space. This does NOT excuse or dismiss Richard Stallman's explicit endorsement of misogyny, pedophilia and sexual assault. If the FSF wants to continue a notion of being a safe intersectional space, RMS has to be removed. Full stop.

I don't know much about issues and complications within the OSI but no place is perfect or devoid of things if they haven't been discussed or addressed. It'd make sense for them to employ a company culture team to weed it out in advance.

Where Does the Future of F/LOSS stand?

To shortly answer Kat's question, I think that the FSF itself does not stand to help, support or reach out to marginalized people in the ways that they claim to by endorsing and employing people who actively cannot live to those values. The concept of free software isn't a novel one either, there's facets of it that can be seen in indigenous cultures (the Taíno and Arawak immediately come to mind). But as they stand, they force a lifestyle and behavior that people of color (myself included, tbh) can find extremely difficult to uphold and be ostracized for attempting to do so. As someone who does his very best to uphold these ideals because I believe that they can put people first, I don't think that we'll have the best chance to continue the crusade against hyper-proprietary solutions or combating bigoted people if we maintain the status quo. It's not sustainable.

No, Seriously, What Do We Do?

We got amazing people laying out how the visible vanguards of movements are actively harming people in the space (if not directly then by endorsement and support of those who do - it's called being an accomplice). I'm thinking that it's nigh time that we actually give up on things of the GNU. Our endorsement and support is going to further give them a head count that, in this space, is implicit power. And I refuse to give them that knowing that this behavior has been going on for more than a decade.

I think it'd been more advantageous for people of color who have the resources, capability and time to begin divesting from these systems and look to build our own. Greenfield projects are difficult but they can yield more beneficial in the end than spending more energy attempting to refactor a system that's actively resisting changes to grow. This is my personal (unwavering for now) stance about this and that's the answer I'll give in response to Kat's question.

What are You Going to Do?

I have quite a few projects that I've made open and available and I'm already in the process of moving them over to the Parity Public license backed by License Zero. The reasons for this are many but it's largely because OSS as it stands does not make it easy for people without the immense privilege to write, distribute and maintain code to sustain themselves. Just ask the JavaScript community that isn't backed by Facebook, Apple, Google or Microsoft. It's gotten a lot easier due to communal efforts from Debian and corporate endeavors from GitHub. But we need to actually acknowledge the barrier and not make it into a marketing tagline.

The larger community also seems very willing to defend and support people who perpetuate these behaviors to no end. There's threads on THREADS about this on Twitter. Sarah Mei has been telling y'all for the longest. The stuff I'm spending time and energy on is not something I'd want to attribute to abuse and I'd like to optimize to reduce. And I don't have the energy to explain to people anymore why they need to go back and read what people have said in the past.

2019-09-13T19:14:50.40215-07:00 • 46 interactions

Clearing Out My Feeds

I'm definitely a very "connected" person. After an hour or so of being up in the, I'm checking the cached notifications that appear when my phone leaves "Do Not Disturb" mode. It's from an array of platforms (YouTube, Patreon, Twitter, Instagram and mailing lists). I've always kept my subscriptions on YouTube low because the act of watching video is a very isolating operation. Patreon was easy to because it costs money. Instagram is something I'm working to reduce. Twitter, though, has proven to be the most difficult platform to reduce my "mental traffic" to. I also use it as a pipeline for news.

Thankfully Tokimeki Unfollow, a free app on Glitch, has helped me with reducing the number of people I follow as well as moving people into lists that align with the kind of content that they tend to post. I've gone from nearly following 2,000 people to about half of that. And I don't want to stop. I'd like to get to a place where I don't see updates if it's not a mention or a direct message. And even then, I'd like to direct those to my personal site's contact form. Once I get down to about 750, which feels like an optimal number of people to follow for me, I'll double down on adding this content into Lwa so I can just follow Twitter lists like a timeline - instead of relying on the primary timeline for the source of information. I'm excited to launch it!

2019-09-11T20:31:32.94994-07:00 • 3 interactions

Tags:
silos
twitter

Koype 0.1.0 is Out!

Friday evening, after some testing of the development environment, I finally cut the first minor version of Koype. It's been nearly a year since I've started working on this project and this minor version reflects a shoddy attempt at a self-service IndieWeb site. There's a lot of implications and assumptions I've made and I'd love for those willing to try it out on their systems to help me break them. Some of those are:

You'll be using an object storage system to store all of the files in the system. Koype stores information about its posts, incoming Webmentions and media all to object storage. This can be amended to work with just local file storage given demand and interest for it comes into play.
You'd use SQLite to store information about your system. I almost went with using PostgreSQL because my familiarity with it but I decided to give SQLite another look. After noting that it's already basically everywhere and moves well, I figure it'd be something I would want to stick into Koype to test it out.
You have some level of familiarity with Elixir. When things break, there's a console you would want to drop into to push things back into the normal state. I'm working on a way to build interfaces for this moments instead of writing those commands over and over.

I'm super excited to see how y'all go about using this initial version. I'll be bringing more bug fixes to Koype but will be turning a bit more attention to getting things around Koype like the Hub and Theme Gallery up and running. You can install Koype today by using its Docker image on Docker Hub or building from source. There's more information coming to the docs so check for an update on that soon!

2019-08-30T21:27:56.59918-07:00 • 19 interactions

Tags:
releases
koype