Emily Post for the digital generation.

“Geek to geek communications” with Michael Schwern

barcampblock logo

I’m at BarCampBlock so there’s going to be a bit of a change in format here while I liveblog Michael Schwern’s talk, “Geek to Geek: how we fail, how to fix it.”

What’s this talk about?

  • We’re good at computer science and bad at people
  • We have certain patterns of failure and certain ways to fix those
  • Schwern is especially looking at mailing lists and bug reports
    • Mailing lists are where we mostly talk to each other
    • Bug reports are where we ask each other to do things, and where we start to interact with “real” users
  • There’s no existing clearinghouse for this information
  • Schwern completely missed GeekEtiquette! So what other resources are out there? The conversation is open.

Going round the circle, people introduced themselves and talked about why they’re interested in g2g communications.

  • geek communities like IRC, how to avoid toxicity
  • internal technical documentation (person who raised this works at Google)
  • interfacing with business/sales people
  • QA and bug reports and how to smooth the process

Literalist fuckheads

So, let’s look at what geeks are like. In general, we’re:

  • analytical
  • literal
  • very direct
  • truthful - academic purity
  • aggressive

Schwern gives a quick anecdote about what he calls “Literalist fuckhead mode”. When he arrived at BarCamp, he ran into someone and said, “Excuse me”. The other geek said “No.” “Huh?” said Schwern. “Oh, i just like to screw around with the social conventions,” said the geek.

That’s a geek fallacy right there — the idea that the social niceties must be questioned. Words aren’t necessarily meant to be taken completely literally, they’re communicating something else: respect, etc. (We come back to this later when talking about Tact Filters.)

So, on the subject of literalism, Pete K raises the sort of communications you see with literalist/geek kids:

Parent: “Could you get the stuff from the car?”
Kid: “Yes.” (You didn’t ask me to actually do it, just whether I could.)

Parent: “Why didn’t you get the stuff out of the car?” (Rhetorical question, intended as a prompt/reminder.)
Kid: “Because… blah blah blah.” (Responds to literal question.)

Sometimes when kids do this it’s just because they simply parse things literally and honestly don’t realise the social conventions. In other cases (one participant pointed out) it’s because they’re being annoying on purpose. The “excuse me” literalist was probably in the second category.

Is it time to get past our Aspergers’ issues?

All this — the list of geek attributes — is low-level aspergers stuff. That puts a label on it but doesn’t actually help us communicate and get stuff done.

There are a lot of people here — in the session and in the wider geek community — who have a feeling that “something is wrong”.

At OSCON they had a “people” track and Schwern wasn’t the only one talking about this. Lots of people are realising there’s a communication problem but don’t know how to deal with it.

Well, says Schwern, we’re smart and analytical. If we can find out what the problem is, we can try and figure out what it is and how to fix it. So for now, let’s try and identify the anti-patterns in geek communication.

The problems of textual communications

One common problem with g2g communications: text. I can’t even demonstrate schwern’s “fuck you” act here because it had lots of non-verbal components. ASCII has no context. ASCII has no empathy.

We have emoticons and actually they sort of help. But aren’t emoticons a band-aid for unclear conversation? People who write professionally feel that emoticons are a sellout, because you should’ve expressed yourself more clearly in the text itself. Geeks sometimes like emoticons because they achieve the same effect as a paragraph of words in 2-3 ASCII characters.

Besides, we don’t write like pro writers, crafting our words to perfectly express our meaning; instead we transcribe what we’d say out loud, and we type fast and don’t stop to think.

In email, some of us suffer from “Me too” damage after years on Usenet. We don’t like to reply unless we’ve got something substantial to say. But this can lead to Warnock’s Dilemma:

The problem with no response is that there are five possible interpretations:

1. The post is correct, well-written information that needs no follow-up commentary. There’s nothing more to say except “Yeah, what he said.”
2. The post is complete and utter nonsense, and no one wants to waste the energy or bandwidth to even point this out.
3. No one read the post, for whatever reason.
4. No one understood the post, but won’t ask for clarification, for whatever reason.
5. No one cares about the post, for whatever reason.

The Apache project gets round this using a +1 / 0 / -1 voting mechanism; a quick response saying “+1″ means “I read this and agree,” -1 means “I read this and disagree”, and 0 means “I read this but don’t have any particular opinion.” This works well if everyone understands, but if you start doing it at random on a new mailing list it might cause some confusion.

Instant messaging

Sometimes IM is a better communication method than email. Even though it’s text, it provides some non-verbal cues. For instance, interjections, time delays, and notifications that the other party is typing can all provide additional context.

“I’m better in text because I can backspace, delete, etc.”

“Backspace only works until you hit return.”

One common IM/IRC problem is people who ask you if they can ask a question. Ask-to-ask is a mainstream cultural behaviour that appears polite in most circles, but geeks find it inefficient. On #perl on IRC there used to be a bot that would reply to anyone who asked to ask. Ask-to-ask is an outbound tact filter.

Tact filters

Most non-geeks have outbound tact filters: they filter what they want to say and add polite noise as it passes through. Geeks have inbound tact filters: they take bare communication with no politeness and just wrap it in assumed politeness as they interpret it.

When non-geeks talk, geeks think the polite sounds they make are redundant.

When geeks talk, non-geeks just think they’re being incredibly rude.

Pete’s son at age 6 went through a period of hating tact words (please, thank you, etc) and wouldn’t say them and got upset when people said them.

Tact filters in email

Schwern and I both recently started using email salutations/endings as outbound tact filters. Typical geek email doesn’t have saltuations/closings, but now we’re saying “Hi, Stan,” and “All the best” and the like.

I got this stuff from Send: The Essential Guide to Email Etiquette which I described as “the geek’s guide to email camoflage”. Mostly it’s aimed at business people and doesn’t fit geek communication styles, but it does help you get a feeling for what non-geeks are expecting.

Many people talked about “mirroring” with email and outbound tact filters in general: observe what the other party says, and say the same thing back: if they say “Hi, Bob” then you reply with “Hi, Alan.” If you are initiating the conversation, though, you can’t mirror. So you need to have a mental profile of the type of person you’re talking to, and think about what they’d be expecting.

Isn’t that kinda phony?

Is camoflaging or using outbound tact filters being “phony” or somehow untruthful? Typically, if geeks don’t mean something then they don’t want to say it. But building rapport doesn’t actually change who I am. In fact, literalist fuckhead mode can be a form of lying, by blatantly ignoring the non-verbal signals.

Let’s get back to g2g

Schwern tries to get the conversation back onto the subject of g2g communications rather than g2other… with only limited success.

Ken says: what we really ought to be talking about is Aspergers Syndrome. A type of high-functioning autism, aka Geek Syndrome (c.f. Wired).

But is Asperger’s really a disorder? Is treating it that way just “pathologising the other”? US culture seems to love medicalising this stuff and medicating it. (NB: that comment was from me, speaking from an Australian perspective.)

Other people had discomfort with geek/non-geek duality: we live on a continuum depending on what we’re doing at any given time. Some people have geekish tendencies not related to code: music geeks, theatre geeks, etc.

“I like urban planning a *lot*.”

“I think we’ve forgotten it comes from the whole biting-the-head-off-chickens thing.”

“Are we still allowed to look down on those people?”

Other frameworks

One attendee suggests other frameworks to look at: dating advice, relationship books, books about interpersonal communications. All these might provide perspectives outside of the autism-spectrum discussion.

Geek girlfriends vs non-geek girlfriends. Who here primarily dates geeks? About half. Various personal opinions on how they find it easier to communicate with geek girlfriends/boyfriends.

On empathy

Two components to empathy: wanting to communicate with other people and wanting to talk and listen; and understanding that not everyone is the same, and need to be profiled individually rather than as a member of a set.

“Was there a point where you realised that not everyone else was an idiot, they just didn’t know the things you knew?”

“What? They’re not idiots?”

The thing is they just have a different set of domain specific knowledge. We privilege our own knowledge set. But presumably business and marketing people have their own DSK too. When you get geeks and business people together they don’t understand and don’t care about each other’s DSK. Then each party tries to show off how smart they are by showing off their own DSK and they completely miss each other.

The wrap-up

Schwern’s set up a mailing list to discuss this: g2g@schwerniverse.com

“It’s a mailman list and I’ll just assume you guys know how to deal with that.”

Finis.

Share and Enjoy: These icons link to social bookmarking sites where readers can share and discover new web pages.
  • Digg
  • del.icio.us
  • Reddit
  • Slashdot
  • StumbleUpon
  • Technorati

17 Comments so far

  1. […] Geek 2 Geek Communications with Michael Schwern: […]

  2. Chris Messina August 20th, 2007 12:24 pm

    This is a great post and summary write up. I’m sad I missed this session — thanks for recording it!!

  3. […] The discussion included tact filters, Warnock’s dilemma, and the literalist fuckhead syndrome.read more | digg […]

  4. Beatrice M August 20th, 2007 4:48 pm

    I ditto Chris’s message. These notes are very well done and I would have liked to have listened to this talk. We were having a conversation much like this talk, this morning. Communication is hard, but luckily can always be improved upon.

  5. Mary August 20th, 2007 6:04 pm

    Hrm, is my comment still in moderation, or has it just gone?

    In any case another observation: it helps to think of a lot of behaviour in terms of goals and ongoing relationships. You may, in any particular interaction only have a goal of exchanging information relevant to task X. This is very geeky. Other people usually have additional goals, which include any of the following: feeling good about themselves, making the other people feel good about themselves, making the other person feel bad about themselves, making sure the other person remembers them as someone smart/accomplished/considerate/superior so that future interactions will be to their benefit… all kinds of things. That’s the padding in most conversations.

    It’s game theory, with all kinds of extra subterfuge where they know you’re trying to look considerate so you have go beyond the bounds of consideration called for in this conversation in order to seem really considerate.

    It’s difficult to model someone as “wants all conversations to be pure information exchange, will look down on me for anything else” very quickly. (And really, of all the people I know who answer ‘or’ questions with ‘yes’ — folks, English has a different grammar and semantics to first order logic, not a necessarily inferior one — I’ve never met one who didn’t at least have the secondary social goal of establishing either “I’m geekier than you” or “I care less about you than you do about me”.)

  6. Skud August 20th, 2007 6:59 pm

    Mary: just gone, I guess. You’re off moderation — only new posters are moderated.

    I love your analysis of the or -> yes question/answer technique. Though if done between geeks who know each other well, it doesn’t necessarily have that “I’m smarter” thing, but might just be a mildly humorous response to maintain geekish rapport.

  7. Aristotle Pagaltzis August 20th, 2007 8:18 pm

    Skud: and that’s exactly what Mary is saying – maintaining geekish rapport is a social goal. :-)

  8. Mary August 23rd, 2007 7:20 am

    Skud: I think I’m biased against answering “or” questions with “yes” mostly because when someone does it to me, I’m invariably in the (more truly geeky!) position of trying to get information from them fairly quickly!

  9. Aristotle Pagaltzis August 24th, 2007 7:22 am

    I’ve now gotten around to actually reading the entire writeup. Thanks for the detailed notes!

    I want to add a remark on a specific issue: ask-to-ask being disliked in chat rooms isn’t merely a matter of processing things with an inbound tact filter. There are a number of ways in which “physical” constraints and social patterns of chat rooms combine to make this behaviour actually impolite. Chat rooms are asynchronous and tolerant of concurrent conversations, which leads to participants operating on partial (or really: periodic full) attention.

    This is exactly the opposite of communication using audible speech. When you want to speak to someone, they can listen to no one else meanwhile, and they will have to concentrate on you. Requesting permission for their attention rather than blithely hogging it is polite. In a chat room, most users are occupied with something other than chatting; those who are available to answer are by definition already focussed on the room with their scarce full attention. Asking them whether they are willing to answer is a waste of that attention.

    Just as it’s rude to hog someone’s attention who isn’t willing to grant it, it’s equally rude to waste it when they’ve already given it. When IRC folks say you shouldn’t ask to ask, it’s not a manifestation of geek directness so much as a result of how chat rooms work.

  10. Sam August 25th, 2007 4:43 pm

    This reminds me of something I wrote awhile back - you go into more detail on a lot of it here, and I’d have loved to be at that talk, but I’ll pass on the link in any case - http://mirrorshard.livejournal.com/80232.html

  11. Tiffany September 4th, 2007 10:08 pm

    There are some really interesting (and true to life) points about geek communication in those notes. Thankyou for bringing them to us.

    If you have time, I’m a non-computer programming geek, but I like working with all types of geeks, while retaining a fair ability to communicate with the real world. It seems to me there is plenty of need for people to form a translation service between computer programmers and the real world. Even if communication can be improved, many geeks are going to learn social things slower, and need someone to explain social etiquette to them. And there is something refreshing about using geek styles of communication in highly technical contexts - it’s so much quicker at passing on information. (It’s choosing when and with whom to have such conversations that is an important social skill).

    My lateral question, should anyone have spare time, is which interface jobs exist between geekdom and the rest of the world?(Or more specifically computer geekdom, as a concentrated nexus of paying geekdom). Which ones are open to people with no programming training?

  12. Anna September 11th, 2007 3:34 pm

    Aspergers is more than just being a geek, unfortunately. And there’s no drug for it.

    However, Aspies (and other non-aspie geeks) can (usually) consciously learn to apply certain “social niceties”. It’s just a conscious effort rather than picking it up by osmosis like neurotypicals do.

    FWIW - I LOVE the description of incoming/outgoing tact filters. That’s really helpful to read about. Thanks. I just wish I’d attended this talk. It looks like it was great.

    My lateral question, should anyone have spare time, is which interface jobs exist between geekdom and the rest of the world?(Or more specifically computer geekdom, as a concentrated nexus of paying geekdom). Which ones are open to people with no programming training?

    I’m also interested in reading the answers to this.

  13. […] asked the following question in the comments to my earlier post about Michael Schwern’s Geek2Geek Communications talk: My lateral question, should anyone have spare time, is which interface jobs exist between geekdom […]

  14. […] read more | digg story […]

  15. […] post about geek communication I recently came across this blog post summarising a talk by Michael Schwern at BarCampBlock about communication among geeks…. I recommend it to all, geek or otherwise, as it makes some interesting points about how geeks tend […]

  16. Am I a geek? « MBSProgress January 10th, 2008 6:58 pm

    […] , Sales , Speaking , communication , public speaking An absolutely FASCINATING article here. Called Geek Etiquette, it is a summary of a talk by Michael Schwern “Geek to geek:how we […]

  17. stratobiker January 27th, 2008 2:35 am

    How is a ‘geek’ defined for purpose of study?

    I thought geek was part of a continum.
    SB

Leave a reply