My feed reader service is https://feedmail.org. The simplest instructions for WebSub are those on the public Google hub https://pubsubhubbub.appspot.com/. But there are other public hubs available or you can run your own.
The worst use of the <BLINK> tag ever was the discussion held in the early days of RSS about escaping HTML in titles, whose attention-grabbing title went something like this: "Hey, what happens when you put a <BLINK> tag in the title???!!!"
The content of that notorious discussion went on and off and on and off for weeks, giving all the netizens of the RSS syndication community blogosphere terrible headaches, with people's entire blogs disappearing and reappearing every second, until it finally reached a flashing point, when Dave Winer humbly conceded that it wasn't the user's fault for being an idiot, and maybe just maybe there was tiny teeny little design flaw in RSS, and it wasn't actually such a great idea to allow HTML tags in RSS titles.
2.0.1 was published in 2003 and I don't expect to see a 2.0.2 or 2.0.3.
The ambiguities you mention are real, though. “The url must be an http url” is one that feed validators would adhere to very strictly, for example, rejecting HTTPS URLs as invalid[1], which even caused Apple's Podcast submissions process to reject feeds with HTTPS URLs in them at one point (I was employed at the time by a company that hosted podcasts, and we had to work around it by rewriting URLs to use HTTP).
I received a reply from the spec's author to my email suggesting they update the spec to clarify this (“of course you can use HTTPS”) but the spec itself was never updated and I would consider it unmaintained at this point.
Atom and JSON Feed are good alternatives. Atom because it has fewer ambiguities and JSON Feed because it has an open GitHub repo[2] and was updated in 2020.
I recommend that people use Atom for new implementations for basically this reason. However to be honest it probably isn't a big enough problem to switch to Atom of RSS is working for you.
I was intrigued by the title of this article and if I remember well Atom was a sensible standard that was quite well defined, but specifically Dave Winer didn't want it to happen because it was not invented by him. I switched all RSS-y stuff I had to develop + all sites I control to Atom long ago, and it pretty much just works.
I was wondering the same. At some point it appeared atom was going to take over the feed space, but when I started blogging again last year it appeared we had more or less standardized again on RSS 2.
> a user did not have to find a rssCloud server for their feed to reference – their own site could handle feed registrations and notifications to feed reader apps. This created a supply of literally millions of weblogs that could support quick notifications to their readers if their reading app supported the use of rssCloud.
Um, is it just me or does that sound like a DDOS waiting to happen? (I'm out of the loop on these things.)
I'm not familiar with RssCloud but WebSub requires the notification endpoint to confirm a subscription before sending new entries. So your DoS amplification is one small HTTP request for a small HTTP request. I guess at best it would make it slightly harder to block as it is coming from everywhere.
Interesting comments about Dave Winer in the article.
I first became aware of him in the last year or two via a mention here on HN. Since then, I have been receiving and reading his daily blog post via email. (He claims his blog has been running longer than any other in existence, could be true). They guy is clearly smart and hard working, and has lots of ideas, many of which seem good. What he shares about his personal life is also interesting, I could imagine being friends with him if we were in close proximity.
However I also get the sense sometimes that he is a "legend in his own mind", to the detriment of the very technical causes he is promoting. He is fixated on writing tools such as "outliners" (MS Excel and Word have all the outlining features I have ever needed, which isn't very often), and whether or not RSS items need to have titles. The things he agonizes over are, I think, just background noise for most users, and his illustrious history in a niche area of Apple desktop computing decades ago doesn't really buy him anything in today's marketplace, to his (unstated) frustration.
Dave was really a pioneer. In the early days of the Mac, he created an outliner called MORE which was a very innovative product for the time, which I immediately loved and used as the center of all my information-keeping. He sold it to Symantec for, as I understand it, a couple million dollars (probably twice that in today's money), but then it disappeared. (OmniOutliner, made by people with no relationship to Winer that I know of, carries on MORE's tradition.) Then he made a product called Frontier, which was a programming environment that never really took off. Then when the internet got critical mass he came up with RSS and that was obviously very useful and influential.
His claim to have had the first blog is probably true. I remember reading it way back then.
He's a bit of a prickly character, always has been.
And before MORE, he created Think Tank for PC and Apple 2.
I remember reading him say his first outliner was on a timeshare system in the 1970s, and he took inspiration from that Douglas Engelbart’s Mother of All Demos
This seems a little unfair, maybe unintentionally on your part. If you’ve been reading his blog long enough (and I don’t know the value of “long enough” since I’ve been reading so long) it becomes clear he situates new proposals and tools in into a ton of context - how he came to think this new thing would be useful. This involves discussing history. By necessity this disproportionately is his own professional / independent developer history. I find this approach generally refreshing and honest - most software emerges from a path dependent individual context but people tend to suggest they just developed the objectively best solution to a widespread problem. Dave tends to be more honest about scratching his own itch and the influence of his own history on how he approaches problems.
This approach can read as ego / bragging maybe, if you’re not familiar with it. “Why is he talking about his past projects so much,” etc. Well because it led into why he thinks the new project is a good idea.
Not to say he doesn’t have some ego like all people. He does want people to know for example that podcasts were a natural outgrowth of RSS, so natural he hand a hand in adding it himself despite not having an audio background. But his point is almost always that podcasting should stay open, not that he is some genius. If anythint he tends to encourage non developers (eg journalists) to be more involved in developing new standards.
(Also the “fixation” on blogs without titles is about the fact that RSS the standard is perfectly capable of carrying Twitter type items without titles, since it only specifies either a title or description as required - but feed readers mostly assume titles which has encouraged a heavyweight high effort concept of what a post is, to the advantage of closed platforms like Twitter and to the disadvantage of the RSS ecosystem. I think he has a very good point on this. These details matters. That they are “background noise” to many users - well so are most important technical details. Try getting most users to understand the difference between transport and end to end encryption for example. )
> He does want people to know for example that podcasts were a natural outgrowth of RSS, so natural he hand a hand in adding it himself despite not having an audio background.
In that he is right by the way: He had a back and forth with Adam Curry, I believe, about audio attachments, and as a result Winer added the enclosure element. Everything else about podcasts, the name, the MediaRSS extensions, the coopting by Apple, the iTunes directory and extensions, the professionalization, came later. Funnily enough Winer tends to think about his own podcasts more as a stream of consciousness voice memo than an edited product.
(Yes, I'm reading him that long. Like pretty much everyone, who reads him that long, I have mixed and complicated about him.)
> MS Excel and Word have all the outlining features I have ever needed, which isn't very often
This says more about you than it does about the usefulness of outliners. Not that that's bad, but it's important not to classify them as trivial just because you don't use them. Outlining software is exploding in popularity. Some examples include org-mode, Workflowy, Dynalist, Roam Research, Logseq, and Remnote. It's also fine to dislike DW, but it's not accurate to diminish his importance because you don't personally use outlining software.
Outlining is a crucial step in authoring most anything with substance, from essays to books and even project plans (most gantt software Will collapse subitems like an outliner)
My Guess Is mr Winer writes more than you do and therefore uses more outliners.
Besides, your argoment Is a major ad hominem, which makes a bit invalid.
I included the information about Dave Winer in the article because (1) there have been problems in his products regarding rssCloud support for WordPress sites, which I wanted "on the record", and (2) I felt that his attitude toward "help" needs to be called out, otherwise people would not be aware, and have that context in reading his blog/side of the story. He mentions Automattic in his post on December 24th (http://scripting.com/2022/12/24/162940.html#a163527) regarding rssCloud work with WordPress, but it was really Andrew Shell (http://feedland.org/?item=880552) that did the work to make it happen.
Now, we can debate all day long about how/why/what, but it really doesn't matter. He is who he is and nobody is going to change that. The guy must be in his 70-80's at this point. After all these years, he's still treating people weirdly, it must be a personality tick of his.
We can decide if we want to work with him and his stuff, or not.
"I've made a point of reaching out to old friends in the last few weeks to see if we can work together on any projects. So far most people are still wary of working with others, it seems." -Dave Winer
We can think of him like we do Steve Jobs (and based on his ego, that's my guess of what he wants). He built something that got people's attention, but treated people poorly in doing so.
We can also decide how we want to act towards others. Don't be like Dave.
What is rssCloud exactly and why should a long-time blogger care about it? I ask this not to be facetious. It's because I read your blog post with interest, but by the end, I still wasn't sure. Your second paragraph, which defines rssCloud, gave a kind of technical description (the "what") but not the user benefit description (the "why"). Unfortunately I found the Wikipedia article to be much the same. Basically, I couldn't really figure out "what's in it for me". For context, I'm a long-time programmer, I have blogged at amontalenti.com since ~2004, via WordPress since ~2005, and I even work for Automattic. So, if I'm confused, I bet a lot of people with less context are even moreso.
The main benefit of rssCloud is quicker notification of updates versus polling every 10 to 60 minutes, as most feed readers do. The drawbacks are that the tool generating the RSS feed needs to include another element (the cloud element), the tool needs to inform a server that an update has occurred (need a server), and there has to be an app that registers with the server to get updates, and to display those updates (need an app that supports rssCloud). Without all of those 3 pieces, rssCloud is not going to provide the value of quicker notification. You could also consult your fellow employee, Joseph Scott, who worked on the WordPress plugin, for more context.
I am not familiar with what is needed to support WebSub, but I assume there is some work involved, just as there would be for XMPP, Jabber and any other protocol that involves more than polling. That "extra work" is the barrier to entry that helps keep people using Twitter and Facebook - the users get that notification "hit" and network effect without any extra work on their part.
Thanks, that makes sense. So the core problem it solves is preventing the need for polling for feed consumers and apps, allowing for real-time notifications of updates. In theory, people could use rssCloud to adapt blogging platforms into microblogging use cases, e.g. using the simplicity of an analogy, Mastodon use cases built atop WordPress infrastructure.