ACK vs. NAK

Peter Rukavina

When we are sending messages — messages of any sort, be they SMS, morse code, messages in a bottle, waving hello across a crowded street — we need a protocol, a set of expectations for how the exchange of messages will play out.  Key to such protocols are two types of protocol messagesNAK and ACK.

NAK, or “negative acknowledgement,” is a protocol message sent when something’s gone wrong. “I can’t hear you.” “Noise on the line.” “I don’t understand.” “Please explain.” 

ACK, or “acknowledgement,” is a protocol message that signifies ”okay, got that, next…”

NAK and ACK can be used together, but, more often than not, their use anchors two fundamentally different methods for communicating.

Think of giving your credit card number over the phone. You can simply rhyme out the digits — “4111 1111 1111 1111” — and wait for the other party to interrupt or to reply, when you’re done, “could you repeat that.” A NAK, in other words. Or simply to proceed. Which allows you to assume they got the number.

Or you could spool out the credit card number digit by digit, waiting for confirmation from the other party. “4” (they say “4” or “OK”), “1” (they say “4” or “OK”), and so on. ACK ACK ACK.

These protocol messages are not only part of human communication, but also part of the very lifeblood of the Internet upon which you read this post. It comes to you, down in the engine room, via TCP/IP, a “reliable, ordered, and error-checked delivery of a stream of octets between applications running on hosts communicating over an IP network.”

Which is to say, a way of getting these words from my server to your browser in a way that works out for all concerned.

In the human realm the messaging systems we use most often, outside of face to face communication, are email and texting, and in both of these there is a fog of confusion surrounding whether we’re expected to use ACK protocol messages or NAK protocol messages.

I notice this fog most often with generations younger than mine: I will send a text message or email and never receive a reply. I’ve deduced that this isn’t because the message isn’t received, nor because it isn’t read, taken to heart, acted on. It’s simply that younger people than I have switched to only sending NAK when needed, and never sending ACK: they’ll let you know when there’s noise on the line, but otherwise it’s faith-based communication.

The contrast with my generation, which is more ACK-based, is more palpable because many of my peers over-ACK. So they not only reply to messages that cry out for an ACK, but they will often add a coda message, something like “Have a Great Day!” or “Thanks for doing this!” or “Great.” The youngers consider this, from the first ACK onward, wasteful over-communication.

As I sit almost exactly in the DMZ separating the ACKers and the NAKers, I have the most confusing position of all, as I need to decide, on a message by message basis, which way to go. And, especially in recent years, as I try to trend younger and fight off my impending agedness, I opt for NAK. Which is a mistake when there’s a clear expectation of ACK.

This all came to a head in March during a protracted, stressful flurry of ACKnessless for which I was socially called to task.

As a result, I’ve jumped over to the other side, going so far as to not only always ACK, but to prophylactically establish an expectation of impending ACK by engaging in pre-message messaging. “I’m heading to the train station now. I will text you once I’m on the platform and then, again, when the train arrives.” That sort of thing.

If you’re going to commit messaging faux pas, I ration, it’s better to err on the side of being annoyingly over-communicative than annoyingly under-communicative.

Please acknowledge that you’ve read and understood this post.

Comments

Submitted by alexander on

Permalink

This post tells me you'd really enjoy reading The Information if you haven't already. The intro chapter especially has a good discussion of information density vs reliability of human communication. http://www.amazon.com/Information-History-Theory-Flood/dp/1400096235

Submitted by Oliver B on

Permalink

Funny, earlier today I was going to comment how I missed the "Like" functionality of Facebook when I link out to read something on your blog, and a Like is a kind of ACK. Anyway, I've been consternated by the same uncertainty and thought there might be a generational trend. I wonder if transmission efficiency and accuracy determine the system that makes the most sense, or if it's arbitrary and cultural? It does strongly suggest you have complete confidence you'll be heard correctly, not to expect an ACK. Maybe the modes of communication we engage in these days justify that confidence, while the ones in our formative years (or during the years of the parents who formed us) didn't justify it. Actually on my iPhone the bubble that contains my outgoing text shows a "Read" (ACK) when it's been received and read. To the extent ACKing becomes automated, I suppose we're liable to lose the habit ourselves.

Submitted by Oliver B on

Permalink

Your credit card example might be the simplest case--all digits, all expected to be digits. Then there's trying to communicate your potentially alphanumeric email address vocally over a bad cell phone connection, say to somebody at customer service who you'll never be able to reach again if he or she gets the address wrong. Then it's "A as in 'alligator', the digit one..." character by character and you still want them to recite it back to you.

Submitted by Oliver B on

Permalink

I skipped a step: After they read you back your email address, they ask you for an ACK or NAK. It's a hot mess of ACK-NAK.

Submitted by Daley on

Permalink

This is an interesting observation. I also experience this with groups of colleagues who always use the phone which then requires an explicit text-based ACK (SMS or e-mail) to confirm their request.

Submitted by Peter B on

Permalink

Love this observation! Also, I noticed that I (over?) use Twitter faves to ACK comments/replies on Twitter. Fave = got it, thanks!

Submitted by Dave on

Permalink

Love the article - I've even sent and ack though via another channel as well (redundancy; another topic?)

To add to this discussion, I believe that many of us 'ack-ers' were early adopters of technology where we didn't always have always-on connectivity so it was easier to acknowledge the message instead of waiting for the retransmit to follow the TCP/IP analogy. Todays messengers assume assured clear-channel communications so they default to the assumption that the message was received.

Submitted by Pete on

Permalink

The lack of ack or nak in Internet-mediated social correspondence is sometimes called "Warnock's dilemma."

Non sequitur: "Ack!" is something Bill the Cat says.

Add new comment

Plain text

  • Allowed HTML tags: <b> <i> <em> <strong> <blockquote> <code> <ul> <ol> <li>
  • Lines and paragraphs break automatically.

About This Blog

Photo of Peter RukavinaI am . I am a writer, letterpress printer, and a curious person.

To learn more about me, read my /nowlook at my bio, listen to audio I’ve posted, read presentations and speeches I’ve written, or get in touch (peter@rukavina.net is the quickest way). 

You can subscribe to an RSS feed of posts, an RSS feed of comments, or a podcast RSS feed that just contains audio posts. You can also receive a daily digests of posts by email.

Search