4 ms·

JSON Feed

13 points by edent 5y ago

GrumpyNl 5y ago
From the site, "We — Manton Reece and Brent Simmons — have noticed that JSON has become the developers’ choice for APIs, and that developers will often go out of their way to avoid XML. JSON is simpler to read and write, and it’s less prone to bugs." And now they are getting close to XML with all the addons etc. Keep it simple.
- oneeyedpigeon 5y ago
  What changes? The latest JSON spec is from 2017 and it's essentially the same as it was in 2001. People may choose to use JSON to store data in novel ways, but that doesn't create "addons" to the language.
abetusk 5y ago
RSS feeds in JSON [0]. Personally, I think this is great but I doubt very much it will catch on. For anything XML related, I convert to JSON [1] and then parse with jq [2], including (especially) my own personal RSS feed. My opinion is that JSON fulfilled the promise of XML (ubiquitous, self document human readable data) but it came too late for many standards that used XML and now we're kind of stuck with it. JSON has it's problems (not streamable, not easily commentable, any schemas needed are tacked on, etc.) but it's so much easier to work with than anything XML related. [0] https://www.jsonfeed.org/version/1.1/ https://www.jsonfeed.org/version/1.1/ [1] https://www.npmjs.com/package/xml2json https://www.npmjs.com/package/xml2json [2] https://stedolan.github.io/jq/ https://stedolan.github.io/jq/
- detaro 5y ago
  > but I doubt very much it will catch on JSON Feed was very much a flash in the pan. Big noise when it first came out, few places added it, then silence and almost-abandoned Github repo with suggestions for improvements/fixes being ignored. Visible on the site: 2017, then first small update in 2020. That IMHO killed any chance at momentum.
  - bmn__ 5y ago
    It was clear in the way the authors responded to both criticism and improvement of the spec that they would be even worse stewards than Winer, Crockford or Gruber. I am glad this had its deserved death before it became entrenched. HN readers, learn from this. When you produce a spec, abandon ego, and cultivate the spec like you would a child.
    - davewiner 5y ago
      Maybe there's still hope for some stewardship of feeds. I have some ideas I'm playing with. I think the key is not leaving the past behind, but also not being limited by it.
      - bmn__ 5y ago
        You were really bad at handling the RSS spec. Your boundless ego got in everyone's way, and in the end you left everyone disappointed: the feed subscribers, the Web authors, the library implementers. Shame on you. It was so bad that people got together and made a whole new set of specs because it was literally impossible to salvage RSS.
        davewiner 5y ago
        hmm. RSS worked pretty well for something that needed salvaging, as you say. the mess was created by vendors who wouldn't work with each other. i stepped back after 2.0, people could've done anything they wanted.
        bmn__ 5y ago
        Ah yes, blaming someone else, when we have have historic records that it was you who is responsible. Who do you hope to make an impression on with your lies? You haven't changed one bit. • http://p3rl.org/XML::RAI#DESCRIPTION http://p3rl.org/XML::RAI#DESCRIPTION • https://web.archive.org/web/2004/http://diveintomark.org/archives/2004/02/04/incompatible-rss https://web.archive.org/web/2004/http://diveintomark.org/arc... (For those following along, Userland ⩵ Winer.)
- latexr 5y ago
  > I doubt very much it will catch on. There’s little incentive for websites to change to JSON feeds when their RSS feeds are already implemented, working well, and generated automatically. But several good RSS readers added support for JSON feeds when the format was introduced and that’s all you need for it to be viable; it isn’t important if it “catches on” after that when the flexibility is there. JSON feeds were great for me because I generate my own feeds for personal consumption and can now do so with simpler code. I understand I’m in a minority—most people consume feeds and never create their own—but the larger point is JSON feeds may have already done their job: they exist and are supported if you need them.
thirdplace_ 5y ago
I like the next_url option which points to the next batch of items. Always was annoying that web feeds do not contain the entire history.
- oneeyedpigeon 5y ago
  Agreed that it's a nice feature, but I suspect there are other reasons that publishers don't make full feed histories available, which this won't solve.
- bmn__ 5y ago
  RFC 5005 exists. If a feed does not contain the entire history, that is the fault of the person implementing it.

13 points by edent 5y ago

From the site, "We — Manton Reece and Brent Simmons — have noticed that JSON has become the developers’ choice for APIs, and that developers will often go out of their way to avoid XML. JSON is simpler to read and write, and it’s less prone to bugs." And now they are getting close to XML with all the addons etc. Keep it simple.

oneeyedpigeon 5y ago
What changes? The latest JSON spec is from 2017 and it's essentially the same as it was in 2001. People may choose to use JSON to store data in novel ways, but that doesn't create "addons" to the language.

abetusk 5y ago

RSS feeds in JSON [0]. Personally, I think this is great but I doubt very much it will catch on. For anything XML related, I convert to JSON [1] and then parse with jq [2], including (especially) my own personal RSS feed. My opinion is that JSON fulfilled the promise of XML (ubiquitous, self document human readable data) but it came too late for many standards that used XML and now we're kind of stuck with it. JSON has it's problems (not streamable, not easily commentable, any schemas needed are tacked on, etc.) but it's so much easier to work with than anything XML related. [0] https://www.jsonfeed.org/version/1.1/ https://www.jsonfeed.org/version/1.1/ [1] https://www.npmjs.com/package/xml2json https://www.npmjs.com/package/xml2json [2] https://stedolan.github.io/jq/ https://stedolan.github.io/jq/

detaro 5y ago
> but I doubt very much it will catch on JSON Feed was very much a flash in the pan. Big noise when it first came out, few places added it, then silence and almost-abandoned Github repo with suggestions for improvements/fixes being ignored. Visible on the site: 2017, then first small update in 2020. That IMHO killed any chance at momentum.
- bmn__ 5y ago
  It was clear in the way the authors responded to both criticism and improvement of the spec that they would be even worse stewards than Winer, Crockford or Gruber. I am glad this had its deserved death before it became entrenched. HN readers, learn from this. When you produce a spec, abandon ego, and cultivate the spec like you would a child.
  - davewiner 5y ago
    Maybe there's still hope for some stewardship of feeds. I have some ideas I'm playing with. I think the key is not leaving the past behind, but also not being limited by it.
    - bmn__ 5y ago
      You were really bad at handling the RSS spec. Your boundless ego got in everyone's way, and in the end you left everyone disappointed: the feed subscribers, the Web authors, the library implementers. Shame on you. It was so bad that people got together and made a whole new set of specs because it was literally impossible to salvage RSS.
      - davewiner 5y ago
        hmm. RSS worked pretty well for something that needed salvaging, as you say. the mess was created by vendors who wouldn't work with each other. i stepped back after 2.0, people could've done anything they wanted.
        bmn__ 5y ago
        Ah yes, blaming someone else, when we have have historic records that it was you who is responsible. Who do you hope to make an impression on with your lies? You haven't changed one bit. • http://p3rl.org/XML::RAI#DESCRIPTION http://p3rl.org/XML::RAI#DESCRIPTION • https://web.archive.org/web/2004/http://diveintomark.org/archives/2004/02/04/incompatible-rss https://web.archive.org/web/2004/http://diveintomark.org/arc... (For those following along, Userland ⩵ Winer.)
latexr 5y ago
> I doubt very much it will catch on. There’s little incentive for websites to change to JSON feeds when their RSS feeds are already implemented, working well, and generated automatically. But several good RSS readers added support for JSON feeds when the format was introduced and that’s all you need for it to be viable; it isn’t important if it “catches on” after that when the flexibility is there. JSON feeds were great for me because I generate my own feeds for personal consumption and can now do so with simpler code. I understand I’m in a minority—most people consume feeds and never create their own—but the larger point is JSON feeds may have already done their job: they exist and are supported if you need them.

thirdplace_ 5y ago

I like the next_url option which points to the next batch of items. Always was annoying that web feeds do not contain the entire history.

oneeyedpigeon 5y ago
Agreed that it's a nice feature, but I suspect there are other reasons that publishers don't make full feed histories available, which this won't solve.
bmn__ 5y ago
RFC 5005 exists. If a feed does not contain the entire history, that is the fault of the person implementing it.