Notices by George Dorn (gdorn@social.coop), page 5

@ben I agree in principle, but I need to distinguish between bad things the last developer did from bad things everybody who writes node does, since the latter will likely match whatever tutorials / blog posts I end up consulting.

@ben Yeah, I haven't actually tried running it yet. I'm not sure if I should take a huge node_modules dir to mean it's terrible, or just a real-world project doing real-world things that is more than six months old.

Some brief digging suggests it is using something called Nxus, which appears to be a web application framework, because the world needs more JS frameworks.

So all of the controllers are subclasses of nxus-core or nxus-admin controllers. I guess the project implementation doesn't look utterly terrible, but Nxus is about 2.5 years old which doesn't fill me with confidence.

#nxus ?

I'm maybe inheriting a #NodeJS project, as a part-time paying gig. I don't do node. I do the minimum amount of client-side JS to make something work.

If the project is sane, I'm confident I could skill up on-the-fly. If the project is terrible, I'd be in over my head. I don't have the expertise to judge, though.

If it was Python / Django, I'd be able to tell what I was in for in ten minutes.

Where do I look in a node project? What are best practices? What warning signs do I look for?

@DJWalnut Huh. This reminds me of some of the weirder cutups / mashups that Negativland did. Only goofier.

Nobody:

Me: This joke format bugs me. Nobody said nothing? Isn't this a double negative? Did it start in a language where that construction is normal?

Ah, yes, the Internet Of Owership's #PlatformCoop directory.

https://ioo.coop/directory/

Powered by Google's Firebase.

🤦♂️

@ceejbot This is an excellent talk. You laid out the corrupting influence of VC in FOSS really well.

I'm super excited for entropic and hope it succeeds beyond all hope and dreams. And without that success being a catastrophe.

@tom79@mastodon.social Sweet! I'll be happy to help test for you.

I'm also just particularly familiar with the API because I re-implemented it in Python for #wonambi. So I know what it is theoretically capable of, even if it is sometimes a bit weird about how it does things, like the difference between since_id and min_id, because both are poorly named.

Or, (and relevant to @binder) a command-line client that looks like Dave's Own Citadel.

Really, I should just build a bridge between Mastodon and my RSS reader, and use that to read my timeline. At least then I know I won't miss anything, everything can be sorted how I want, and unread/read status will be explicit rather than inferred.

@tom79@mastodon.social It may also help to know that my internet access is not fast. Doing any extra work to get to the toot I want to read is likely to take 10X more time than somebody on a decent broadband connection.

@tom79@mastodon.social I don't actually care much about the display order, I care about not spending the first half of my short reading sessions loading toots I'm not going to read yet.

The long-press option is news to me; I just tried it and it works for relatively short gaps, but I worry that trying to use it to load three days of toots at a time might either take several minutes anyway or get me rate limited. Loading the oldest 40 toots in a gap rather than the newest would have neither problem.

Finally got around to creating the #fedilab issue for a giant annoyance that eats up over half my daily fediverse reading time: https://gitlab.com/tom79/mastalab/issues/947

I continue to be amazed at how many people assume 'reversed chronological' is the best sorting method for social media. I get wanting to know what's going on right now, but I want to read my timeline like a webcomic, not a firehose I dip into randomly without context. Firehoses are for twitter.

@polymerwitch yeah, ugh. we need a measure to ban all flat taxes.

Regarding #commento: it was relatively easy to set up, and will more-or-less work for my two blogs separated by paths (though Commento also assumes one blog per subdomain and breaks with unhelpful messages if you include a path in your subdomain).

It doesn't do Oauth to Mastodon/Pleroma. There's an open MR to implement that, but it's poorly documented. I'd help, but it's #golang and that bugs me even more than JS.

I also don't know #NodeJS at all and am a mere dabbler in JavaScript, but #Schnack is simple enough and laid out in such a way that seems to work for my Python/Django-trained brain. The one twist is that the client-side code has to be built after changes, which took some time to figure out.

As part of the process, I'll try adding docs to the fork about modifying it; so many #FOSS projects lack even a basic explanation of how to make changes to the code.