... works as a web developer in Hveragerði, Iceland, and writes about the web, digital publishing, and web/product development
These are his notes
@fgtech Bridle is always great.
@petebrown Too true.
@ChrisJWilson It’s the big tech and AI companies that are glossing over these issues because ethics get in the way of making money. See Microsoft’s disbanding of their AI ethics team and Google firing Timnit Gebru. The people who know, and in many case funded the ethics research are the ones downplaying it.
@tkoola Yeah, I’ve had the same thought about how the metaverse was memory-holed.
@artkavanagh Supposedly, yeah.
@mbkriegh Thanks. Much appreciated 🙂
@mbkriegh yup. And since the legal definition of plagiarism is a bit broader than statistical similarity it seems very likely that more than 2% of the images generated by Stable Diffusion would be infringing.
@mbkriegh Spreading the joy of finding good stuff to read. 😎
@annahavron My pleasure!
@gregmoore Yeah. So so many grifters.
@chriskrycho Yeah. I’m much on the same page as you
@KevinMarks Yeah, I'm a huge fan of nunjucks, have used it for a few projects but ruled it out this time for a couple of reasons:
I wanted a template library I could use both on the server and in a worker in the browser. This library is tiny compared to nunjucks or even Eta, a similar library I benchmarked it directly against.
Also, most string templating libraries either require support for eval or that you precompile templates if you want maximum performance. That's an issue if you want to keep the option open of either deploying to Deno Deploy/Cloudflare or of locking down the security policy on the client. Precompiling is a pain.
If I were just rendering on the server, I'd use Nunjucks or something like it with no hesitation, though. It's solid.
@pimoore A lot of people are going to recommend DuckDuckGo, but they are using Bing's search engine under the hood, so whatever service deterioration that happens to Bing will eventually migrate to DDG. And, yeah, there kind of aren't any good options out there for search. It's a duopoly where the competing parties clearly have a tacit unofficiall agreement not to make improvements that threaten profitability.
@chriskrycho Oh same.
@V_ Thanks. Yeah, I try 😁
@darby3 So, it seems to have stopped. It just looks like somebody decided to make a full archive copy of a single blog post 1.6 million times, however long that took 🤷🏻♂️
@odd Yeah, probably already the case in some places. At least with a union, they’re likely to have to compensate them for it.
@fgtech That’s a good way of putting it.
@alanralph Pretty much.