Alex Schroeder 🐝 is a user on octodon.social. You can follow them or interact with them if you have an account anywhere in the fediverse. If you don't, you can sign up here.

Mozilla plans to cut RSS feed reader and Live Bookmarks support from Firefox. I can't blame them. It's complex ("outsized maintenance and security impact"). It's the wrong UI ("not supporting any states like read"). I feel we're better served by external tools. ghacks.net/2018/07/25/mozilla-

Another thing that has been bothering me for years, specially when Emacs Wiki was popular, is the architecture depending on regular polling by every single reader. The polling interval must match the posting interval for this to make sense. On a dynamic site like a wiki generating the feed can take a lot of resources if you're relatively small (no caching, CGI scripts, low memory, that kind of thing).

Furthermore, on a wiki, you really don't want the changes of the last 30 days in the feed and if you're going to do that anyway, with full page content, you're wasting resources on the 29 days the user doesn't care about. But how can I fine tune the feed I need to serve depending on the site's activity and the user's visit frequency?

Alternatively, we could have used PingBack or a similar technology that notifies subscribers of changes. That has the benefit of not clogging the net with useless pulls, but there are severe drawbacks: 1. the server must know its subscribers (no privacy), 2. subscribers must be "online" somewhere (just no). That is why we are "stuck" with a technology developers don't really like as soon as demands scale up.

Alex Schroeder 🐝 @kensanata
Follow

And that's why Mozilla can say "RSS/Atom has been in decline and support has been dropped by companies such as Google (Google Reader), or Apple (Apple Mail), or changed focus." And that's why I think we need to keep RSS/Atom despite big companies dropping support because RSS and Atom work precisely for the slow web, the small sites, for average Internet people, for us.

Β· bitblee Β· 52 Β· 50

Feed Discovery works with a tiny bit of Javascript in a bookmark. Here's how I do it, in Firefox:
alexschroeder.ch/wiki/2017-10-

@kensanata live bookmarks was never a particularly good rss implementation, so I’m not surprised Firefox is dropping it. Still, it’s always a shame to hear another company declare that RSS is dying, when its so damn useful

@kensanata Yeah, Google abandoned open source RSS for a reason.

They could not make enough money off it.

They needed something proprietary like Google+, AMP etc.

@onreact @kensanata I don't understand that. It seems to me that Google Reader could've been a superb ad vehicle, as in just like FaceBook people are explicitly telling it what they like.

@alcinnz @kensanata Yeah, but there was potentially a lot of competition.

AMP is entirely controlled by Google.

The competition have to come up with their own tech.

@onreact @alcinnz @kensanata ... it's fair to say that, compared to the amount of users consuming news through Twitter, Facebook, maybe Google+ or (increasingly) WhatsApp or Telegram groups each and every day, the amount of people still using RSS is a *very* small niche. And I think that niche doesn't need RSS supported by Google or Mozilla. But it needs providers still seeing a meaning in providing RSS feeds to a public audience.

@z428 @onreact @alcinnz I agree. I’m sure abandoning Google Reader simply made sense from an engineering perspective: how much effort is spent for what percentage of users. We might think the absolute numbers are large, but capitalism wants Google to spend its resources where the profit is, and sure these engineers could do better. Which is as it should be, in capitalism. Which is why we need alternatives, of course. πŸ˜€

@kensanata @onreact @alcinnz Largely agree here. But I see two "problems" all along the way to be resolved too: (a) We still lack a sustainable funding model for journalism that doesn't have to rely on ads, paywalls or donations. And ...

@kensanata @onreact @alcinnz ... (b), maybe a bit provoking though, I see a load of the FLOSS community obviously almost "agressively conservative" when it comes to open standards and technical solutions, especially talking about use cases outside their own scope. Maybe this is also one reason for the success of Google, Facebook, Twitter, ...?

@z428 @onreact @alcinnz I wonder what you mean by aggressively conservative regarding open standards and technical solutions. Do you have examples? Like defending XMPP vs all the discussion surrounding design decisions made by Signal, for example?

@kensanata @onreact @alcinnz No. More, like, the discussions I repeatedly find myself in about, in example, e-mail-formatting, where still (in 2018) people claim that e-mails *have* to be text/plain with line lengths < 70 chars and aren't even willing to discuss requirements in example by people who want to send out messages that are somehow formatted (colored text, embedded screenshots, ...). Or, as we started right there, ...

@kensanata @onreact @alcinnz ... all the arguing against Atom/APP in favor of RSS, in some aspects completely ignoring that Atom and especially APP covers a bunch of valid use cases some parts of the crowd just aren't willing to accept (APP in example would have made more "portable" creation / posting of online content possible as well). Those discussions seem to often break down to "why even bother, that's just RSS and no one needs more anyway". Or, yet another example is ...

@kensanata @onreact @alcinnz ... the proposed feature of "self-destructing e-mails" in G-Mail. I did work in IT for lawyers in the early 2000s, and I remember having *right this requirement* there for several valid purposes. And I always had people pretty aggressively arguing against, mostly in a manner such as "this is stupid because that's not how e-mail works". I am unsure whether, at this point, Google and other companies manage to win a bunch of users simply by ...

@kensanata @onreact @alcinnz ... accepting and at some point implementing a plethora of just such features. At the very end, it's the same about XMPP, WhatsApp and contact discovery. Still, a load of people in the XMPP crowd claim this is completely inacceptable, and yet a vast majority of people using WhatsApp use it *just because of that* - there is a way to get in touch with your contacts without asking everyone for her or his contact address once you got their phone number.

@kensanata @onreact @alcinnz Possibly a load of things could be better if we were more open to actual end user requirements: How to make privacy-aware contact discovery in a distributed network, in example? How to make (getting back to RSS) a protocol that allows for news syndication *and discussion* in a meaningfully integrated way (quite some bloggers I know went from RSS to Twitter crossposting rather early just because of that)? To be continued... πŸ˜‰

@onreact @kensanata I fully understand that, however. ::( In my environment of techies and people on the #fediverse, there's quite a bunch of people who still adhere to RSS and never actually stopped using it - while I have to acknowledge the vast majority of people in my environment online these days (mostly with smartphones in networks such as Facebook or Twitter) doesn't even know what it could possibly be used for. I really would like things to be different, but guess ...

@kensanata Further, RealTime stuff is just _harder_ - for both producers and consumers. RSS can be a static file on a website. RealTime stuff _requires_ some kind of app running all the time. For both server and client. Much higher bar.

@jos Well said. For me, RSS and Atom feeds allow for the principle of simple things are simple. Real time distribution of notifications can be nice, but there's no simple way to get started.

@kensanata @jos Well, you're in luck, there's a real-time push system part of the RSS2 specification (called rss-cloud) https://cyber.harvard.edu/rss/rss.html#ltcloudgtSubelementOfLtchannelgt

All you need is an internet-facing system which talks SOAP, should be easy to implement a client for your phone.

All jokes aside, this is not as ridiculous as it seems, it was intended for link aggregators.

@Ninjatrappeur @kensanata You lost me at SOAP ;) Now, if we were sticking with old-school, xml-rpc is the name of the game!

@Ninjatrappeur @kensanata Ha, just read the spec --

> which can be implemented in either XML-RPC or SOAP.

DONE!