Why are Internet standards called “Request for Comments”? | Hacker News Reader

Why are Internet standards called “Request for Comments”?

blog.jasoncrawford.org

85 pointsjasoncrawford33 comments

It's interesting and a little ironic to observe how the IETF has evolved over the last 30 years.

Time was, the IETF was more than the standardization effort for the Internet; it was also an intellectual response to the institutional standards body of the day, the CCITT/ITU-T. Where the ITU was bogged down by process, riven by commercial interests and infighting, and unapproachable by researchers, the IETF was animated by "rough consensus and working code".

Clearly, in the contest between the ITU-T (CLNP) and IETF (IP), the ITU-T lost.

Presumably, many hundreds of people were involved in telecoms standardization at ITU-T. Where do we suppose those people went? Did they just give up on their work? Or did they instead migrate to the IETF? Either way: the IETF functions more like the ITU-T today than like the IETF of 1994. "Standards" are owned by denizens of the IETF process; new functionality unknown to the Internet is specified in standards documents before it's ever implemented, or, better yet, "standardized" in opposition to working code.

I'd tentatively suggest that the IETF has served its purpose, and is now at risk of outliving it.

This is an interesting perspective. I always loved the "Joke" RFCs from the earlier days of the IETF. It's sad to see the organization get away from that ethos.

What are your thoughts on the W3C? It seems like they spent the early 2000s going down the path you describe with XHTML, but with HTML5, they've rediscovered "rough consensus and working code," albeit driven entirely by the big browser vendors.

Take my uninformed opinion with a massive grain of salt, but it seems to me that the W3C is mostly glad to leave the HTML5 nonsense to the WHATWG while they work on the stuff they really care about, such as the Semantic Web.

Why "loved" in the past tense? There are new joke RFCs every year: http://www.rfc-editor.org/rfc/rfc6921.txt

HTML5 is the product of WHATWG, which was specifically created to work around the fact that the W3C was trying its hardest to be utterly irrelevant.

> The WHATWG was founded by individuals of Apple, the Mozilla Foundation, and Opera Software in 2004, after a W3C workshop. Apple, Mozilla and Opera were becoming increasingly concerned about the W3C’s direction with XHTML, lack of interest in HTML and apparent disregard for the needs of real-world authors.

Many of them probably went to do similar work at other bodies such as 3GPP - which have managed to produce successful protocols and networks.

I'd tentatively suggest that the IETF has served its purpose, and is now at risk of outliving it.

This matches my own experience, as the author of multiple internet standards drafts and someone who asked them whether it would be possible to open a group to discuss financial networking. What a hornet's nest that stirred up!

Key examples of outdated properties include: finnicky document format requirements (straight from the 1970s), no unicode support within documents (PITA), document character width limitations that can prevent the effective presentation of required information (no fix for this), monolingual nature of its website and resources, and its own bureaucracy despite its official stance: "this is not a bureaucracy".

It occurs to me that we could replace the IETF RFC process with a git repo or a blockchain. Github should probably do this proactively.

I read an amazing article about all that... I think it was focused on TCP/IP, and the big telcos ideas competing. I wish I knew what the link was.

I like how hovering over "kudos", in an attempt to understand what it is, automatically performs an action I didn't want to perform. And there is no undo.

Great UI there, guys. I like how you focus on aesthetic novelty instead of functionality, but I guess that explains why you're hiding the UI all over the site until you hover over crap [1].

[1]: http://en.wikipedia.org/wiki/Mystery_meat_navigation

Yes, yes, a thousand times yes. It's refreshing to hear the voice of reason.

Yep! But u get used to it. though it doesn't mean it's in any way GOOD. Just "pretty".

What does it do?

It's a like button, effectively. Not a Facebook like button, just a 'I like this post' button.

What's really hilarious is they tell you "Don't Move". I made a gif of the behavior in question, so you don't have to try it yourself. I inadvertently gave them like, five more kudos trying to capture it: http://oi39.tinypic.com/24e2kw1.jpg

Sorry, I never make videos, much less upload them, so I didn't really know what services to use. If anyone has recommendations, I'd love to hear about htem.

imgur.com is the current standard in throw away image hosting.

Yeah, but then it's linked to my account then, right? I guess I was looking for more of a throw away picture hosting scheme

No, not linked to your account.

No. If you have an Imgur account maybe, but you could just open an incognito window and upload the picture. No account is necessary.

This is yet another reason to use a Javascript whitelist in your browser.

Add to global stylesheet:

    .kudo.able.clearfix {display: none;}

(Though I agree fully with your comments.)

Considering that Svbtle has been around for like a year and a half now and it's not a particularly annoying UI scheme (rather one with more of a novelty element to it), can we get past debating about the Kudos button? I mean, it's not relevant to the article at all and it's not fair to the author, who had nothing to do with the feature.

That said, so people can catch up:

Here's some backstory on Svbtle: https://news.ycombinator.com/item?id=3742314

Here's Dustin Curtis responding to criticism over it: http://dcurt.is/unkudo

Dustin Curtis' "response to the critcism" in that second link consists of:

1. Quoting a well-articulated email from yet another reader who felt deceived by the trickery; and,

2. (After a preface stating how he finds it amusing that so many people are bothered about this button) The following high-level technical overview of what happens behind the scenes when someone's kudos are volunteered willy-nilly:

  Here’s what it actually does: when you hover over the button, a CSS transform animation is activated which fills 
  the circle. After 1 second (the length of the animation), it fires a request to the server which increments an 
  otherwise meaningless number by exactly one.

This is the full extent of his response. As far as I could figure out, this "response to criticism" addresses the following introductory paragraph of his interlocutor's email:

  I was curious about the Kudos button you mention in your http://dcurt.is/web-standards article. I hovered over it 
  with my mouse pointer, and then it said “don’t move”. I didn’t move. Then it said “sent”. Apparently someone 
  invented a button that does not need to be pressed in order to be triggered.

And totally ignores the sentiment communicated by the more weighty second paragraph which concludes with:

  I very humbly request that you kindly decrement the Kudos statistic by one unit in order to compensate for the 
  one that I definitely did not intend to send until it was forcefully withdrawn from me by a deceitful human 
  interface.

Did Dustin Curtis really miss the subtle (svbtle?) sarcasm of that first paragraph, or am I just imagining things?

Wait a minute, so he didn't realize that the criticism is that it performs a function without me wanting to perform it, and thinks the criticism is that "i feel it sucks out my soul" -_- Seriously, why DID the GP link to this?

Wait, let me get this straight: It's wrong for them to use HN to complain about their bad on-line experiences in hopes of getting people to change software for the better. But it's ok for you to complain about your bad on-line experiences in hopes of getting people to stop talking about their bad on-line experiences.

What makes your experience more important than theirs? If we're going to have complaints here, I'd rather they be towards fixing things than to get people to pretend to stop noticing problems.

That wasn't what I said. All I said is that complaints like this irrelevant to the guy's article. What's wrong with merely asking people to stay on-topic?

Your complaint is also irrelevant to the guy's article. Both you and the person you're complaining about think you have something relevant to the audience. You seem to think that your view of what's good to talk about is more important than somebody else's.

Me, I think the complaint about the bad UI is totally on topic here at Hacker News. I hadn't noticed it before, and was interested to have it brought to my attention. And I agree that it's terrible UI.

Is what passes for a blog post? Copy-pasting 4 paragraphs out of a book?

The Internet is all about free exchange of information... Copyright be darned. :-)

Worse it isn't really even a good copy paste.

I don't even like WikiPedia, and it is a much better explanation on this.

http://en.wikipedia.org/wiki/Request_for_Comments

I consider this the definitive origin story of RFC: http://www.wired.com/wiredenterprise/2012/05/steve-crocker/a...

If you can't beat 'em, join 'em! On your own blog, do a cut-n-paste of what you got from googling a common question, and get the same attention!

One of the things that I love about RFCs is that they're an existence proof.

Sometimes I'll get asked, "How can we have a successful organization without a lot of top-down control of X?" where X is something like architecture or process or coding standards or furniture choice. When people see problems, they imagine solutions pushed through a power structure. And of course, they imagine themselves as the ones in power, forgetting how many bullshit edicts they've had to deal with over the years.

The Internet and its RFCs are my favorite existence proof that you don't need centralized control to get good design and reliable systems. Indeed, you could argue that the Internet, beat out the other early networks because it wasn't centrally controlled.

Makes sense to request comments and input as standards are being developed, but I always wondered why these documents never graduate from "RFC" to "Specification" which is what they ultimately really are.

They can and do become Internet Standards, or STDs.

http://en.wikipedia.org/wiki/Internet_standard

It's interesting that the name "Request for Comments" invokes "Oh, this is a club that I can play in too," as this is the impression I've gotten as I've read increasingly more of them lately and learn how they come to exist at all.

Reply on news.ycombinator.com