@cwebber@garbados I was waiting to hear Diana's take on this! What's interesting to me is that her understanding of Spritely Goblins matches my own basically 100% from all the reading I've done. I am just not able to look at the facts of OCAP and extrapolate "what does this give me *that I care about*?" I'm not saying there's nothing there, just that I am unable to make the leap from "what the things does" to "how will this make people's lives better in a way that current technology does not"
Got into a discussion with someone on Twitter about what makes a programming language "good for beginners". They seemed to want to talk about syntax and stuff. My questions are:
- how hard is it to install - how easily can you import useful data and output results in a useful format - how likely is a beginner to know someone who can help them solve problems they run into - how many resources (tutorials, example code) are out there - can you easily do all this without the command line
People on infosec Twitter keep saying it's extremely bad that lots of people scanned a random QR code. But I'm genuinely not sure how it's different than clicking on a link? My understanding is the flow for most users goes:
- take picture with phone - see url preview - click url
Is the issue that the preview step doesn't exist for a lot of people? Otherwise it seems similar to being presented with any url at all.
@ekaitz_zarraga I just did a little more digging -- https://caniuse.com helpfully has a "date relative" view where you can see a timeline of browser adoption of features. It looks like 2011 was when IE finally adopted all the ECMA stuff that was typically reimplemented in lodash/underscore/etc. But of course what percentage of users were on IE9 at that point vs 8 etc...
I just used Array.prototype.every() and Array.prototype.some() for the first time in production code... ever? I'm sure I could have used them earlier in my career but I always forget they exist. Anyway they are useful functions and I'm glad they're there
If you do a thing in an open source project, and you have a hard time doing it because it's not documented, after you finish doing the thing... document it. If the project has a wiki, put it in the wiki. If it has a docs repo you can write to, put it there. Or at least blog about it so search engines can index it.
Please don't leave the solution to languish in a chat log somewhere.
A year ago I tried learning ActivityPub, and more or less failed. I was confounded by a spec that was so abstract I couldn't make heads or tails of it. Turns out I was missing some key things.
I have written a guide to learning about ActivityPub that I wish existed a year ago when I first set out to learn how to write social media servers that conform to the spec:
@Shamar@amz3@steko I will definitely take a look at GNUnet. Just also want to clarify that I'm hardly a Mozilla person. I'm being paid by them for ten months to do independent research and a lot of my conclusions actually go against their interests (for example, I'm pro internet, not necessarily pro www). This is also why I'm doing intense research into ARPANET: https://write.as/365-rfcs/365-ietf-rfcs-a-50th-anniversary-dive
A while back I built a site that converts RSS feeds to ActivityPub actors that you can subscribe to from Mastodon and other ActivityPub-compliant social networks: https://bots.tinysubversions.com/convert/
I actually built a little toy service (that I will also eventually open source) which converts any RSS feed to an ActivityPub actor that you can subscribe to in Mastodon (or any other AP-compatible client).
Play with it if you like! It is SUPER rough and most feeds end up horribly rendered in Mastodon but it's still kinda cool to see it work: https://bots.tinysubversions.com/convert/
I'm happy to announce that my barebones ActivityPub server, implemented in Node.js/Express, is now open source! I intend for it to be a reference implementation for developers who are adding ActivityPub to their own services, but also it is an application server that you can build on if you want.
In other words, t's a few hundred lines of code that lets you create Mastodon-compatible accounts that can accept follow requests and post stuff to their followers.
I'm the administrator of this server. https://tinysubversions.com is where most of my stuff lives. I'm trying to fix the internet, and some people say I'm at least kind of succeeding. Based in Portland, Oregon, USA. he/him