I go to tech conferences like it’s my job. (Since I do tech advocacy for Cloud Foundry at Pivotal, technically it is part of my job these days.) (2018 update: These days, I’m on the Cloud Developer Advocates team for Microsoft Azure; still lots of livetweeting.) At these events I tweet a lot, to the point where people notice and ask me about it. Sometimes I tell the curious a little about my process; their reactions range from “lolwat” to “you should write that down”. So, here is how I livetweet tech conferences.

What is this livetweeting thing anyhow?

In short, I tweet in near real-time about an event (in this case, usually a conference talk). I think of it as Polaroids: moments for real-time consumption by people reading Twitter, commentary by people both in the audience and not, and promotion to people who could be in the room, if they only knew they should be. Most of my conference talk tweets contain a speaker Twitter handle, the conference hashtag, and at least one photo.

Livetweeting gives me a record of the event for myself, but it also has benefits for the speakers, the other attendees, the conference, and the internet at large. I try to capture and share the mood in the room at that particular moment in time. (A non-zero number of people start following me on Twitter during each conference, at which point I hope they realize that when I get home it’s going to be mostly pictures of cats and gardening and bicycles.)

Why tweet instead of using a conference app or Slack?

I’ll tell you a not-really-secret: I usually don’t bother downloading the conference app, especially if it’s easy to follow the schedule on the website. Yes, I know, you want to encourage community. But guess what? I want to build community and recognition of these topics and speakers in the wider world that will never download your app! That’s why I’m tweeting about the talks instead of writing something in your walled garden. (I’m also not going to auth a conference app to tweet on my behalf.)

Reviews of my own conference talks have included comments about the AC in the room being too cold, the wifi being too slow, or my talk not being relevant to someone’s specific use case. So, talk reviews aren’t my first priority. I’ll leave some if it isn’t a horrible pain-in-the-app, but I’ll admit I don’t always read the ones people leave me (especially if I have to log into something impenetrable). I figure that speakers are going to get a lot more value from public commentary and a lot more people are going to read it.

Slack is fine as an easy way to connect with other attendees at an event, but doesn’t have enough openness to serve as my commentary location. (Yes, I know some Slack teams are archived on the web, but still.)

Speaker Twitter handles

It’s great when conferences collect speaker Twitter handles ahead of time and post them on the conference site, ideally in the speaker’s bio, which should appear on the talk detail page. I’m also a fan of speakers putting their Twitter handles on their slides. Yes, every slide. What if I didn’t take a picture or transcribe it when it was on the screen at first, and now I want to tweet about your awesome talk while it’s still happening (drawing in anyone bored out in the hallway), but first I have to dig up your Twitter handle via some hasty Googling on my phone?

For rapid-fire events with a speaker list that’s easy to check ahead of time (like keynotes and Ignite talks), I am sometimes ambitious enough to create drafts in Tweetbot with the speaker Twitter handles and conference hashtag pre-populated. (As of 2018, Twitter has made the third-party clients less-featureful but I still prefer tweeting with Tweetbot.)

And yes, I realize that sometimes people have their own reasons for not using Twitter. Even if someone doesn’t actively tweet, I like when they register a Twitter account in their name with the profile linking to wherever they can really be found on the internets, because I want to give them credit for their words without spelling out their whole name. Not everyone has a Google-unique name, and if Twitter is auto-completing the wrong handle, a livetweeter might accidentally use that (especially if the right answer was “oh, there isn’t one”).

Conference hashtag

I see the conference hashtag as useful for multiple audiences:

• Attendees at the conference who want to read tweets about the conference
• People not at the conference experiencing FOMO or who want to see what’s going on
• Conference organizers wanting something to retweet, to highlight coverage of their event
• Speakers looking for tweets about their talk
• Your followers not interested in the conference who want to mute a hashtag instead of muting or unfollowing you

Hopefully the organizers have chosen a clear, unambiguous, not-too-long hashtag for the conference and it’s not a mystery and attendees aren’t making up hashtags as they go. And when people have to come up with their own hashtags, Tweetdeck columns can handle the fractal complexity, but Tweetbot on my phone? Less so.

I’ve had people say “I can’t follow you because you tweet too much”, at which point I wonder if they know that the better (third-party) clients have hashtag muting. (And also, it’s really okay if you don’t follow me. It’s not required!)

Side note to organizers: when choosing your conference hashtag, remember, #yourevent is probably fine and #yourevent16 or #yourevent2016 is probably pointless. Are people likely to confuse the tweets from today with the tweets from 2015 or some eventual 2017 version of the event? What problem are you trying to solve by appending those useless numbers? If I’m making up a hashtag, I’m going shorter.

Whether on my phone or on my laptop I always try to have the location set, because it gives people more context if I’m using a hashtag like #devopsdays or #velocityconf that’s used in multiple geographic locations. (When the conf wifi is dodgy and I’m tethering my laptop to my phone, I always need to double-check that TMo or ATT isn’t telling Twitter I’m on the other side of the country.)

Tweet Text

Livetweeting means paying close attention, and it’s a different kind of attention than the absorption of learning. It’s easier to livetweet a talk if I’ve seen the speaker speak before, but even if I haven’t, it means listening both for the quotable lines and for those brilliant insights that are going to be incredibly difficult to fit into a tweet. (I always try to get the speaker’s Twitter handle and the conference hashtag into relevant tweets.)

If I think I might want to transcribe lots of direct quotes, I might use both laptop and phone. Laptop is mostly for “what was the speaker’s Twitter handle” or “let me Google up that repo they mentioned offhand” or suchlike. (Some people use a different style - I like how Jessica Kerr adds marginalia.) (2018 update: My style isn’t the only style, particularly with how picture-heavy my style is. Other styles include far more comprehensive text. I refer you to Liz Fong-Jones and Heidi Waterhouse for excellent livetweeting that differs from mine.)

Structurally, I don’t begin my tweets with the speaker’s Twitter handle, as in that case Twitter is only going to show the tweet in the feed for my followers who also happen to follow that speaker. Some people solve this by prefixing the Twitter handle with a . but I prefer to phrase my tweets about a speaker’s talk so as not to start with their handle. (In theory this is changing at some point; I’ll be interested as to how usable or unusable that makes reading the firehose of an already-active Twitter feed.)

If I’m directly quoting a speaker’s exact words I’ll use “quotation marks” but in many cases I need to paraphrase for length, so I make sure to avoid appearing that I’m quoting the speaker when I’m not. I try to eliminate all unnecessary words (like “that”) while not resorting to “b4 u” for “before you”. If I have commentary of my own, I put it in parens at the end or otherwise make it clear I’m talking about the speaker’s ideas, not repeating them.

Photos

I take a lot of pictures when I’m livetweeting. Sometimes I use burst mode if a speaker moves a lot and is particularly hard to capture, but making selections from burst mode is time-consuming, so I try to get a few regular pics to choose from. I prefer a speaker looking thoughtful without the kinds of facial expressions that come with speech, so I try to catch them in a pause. Hands in view are great, or even full-body if the podium (yes, I know it’s technically called a lectern) isn’t in the way. (Personally I refuse to stand behind anything - I’m just too short, and having most of the room see a tiny floating head isn’t my style.)

I look at the pictures to make sure the speaker doesn’t look distorted or strange - it’s possible to look all sorts of zany (in facial expressions, posture, etc) and I don’t want them to be unhappy about how they look in my tweet. On occasion, a speaker has asked me for the high-res version of a pic I took of them on stage! I keep a staggering number of pictures and am happy to provide them upon request.

Often I take pictures of slides, but I almost always tweet them with another picture that shows the speaker - more immediacy, more personal, more “in the moment”. A picture that can capture both speaker and a slide is great, and I try to choose seating based on camera angles (front-row where possible). And for multiple-speaker talks, it’s helpful when you stand close to one another and look interested when not talking.

If you’re giving a talk and have a great slide you love, be sure to leave it on screen long enough for people to laugh/cry and then grab their phones to take a picture of it! Bonus points if the slide has a large word or two but not a wall of text. As my coworker Coté puts it, corporate slides are word docs in landscape form. I think those are difficult to read from the back of an auditorium.

I usually want at least one picture-containing tweet per talk, which no longer cuts into your 280 character budget, while links, Twitter handles, and hashtags still do. While I don’t particularly want <lj-cut> to be necessary on Twitter, I appreciate now being able to add 1-4 photos without them taking up text characters in the tweet. (2018 update: edited to reflect current reality.)

When I don’t tweet

Sometimes I’m sitting in a talk and I don’t tweet. Does that mean I hated your talk and probably you? Of course not! It’s possible I received a time-sensitive communication from work/family/friends or needed to deal with something that distracted me. Maybe I just ended up retweeting someone else’s excellent tweet instead of tweeting the same thing they already captured so well. Or perhaps I didn’t feel like I had anything to add to the conversation (it does happen!) And sure, it’s possible your talk just didn’t reach me where I am right now, but it’s equally likely I was so engrossed in your well-researched topic or dynamic slide composition or excellent comedic timing that I lost the particular concentration livetweeting requires!

I started livetweeting conferences before I worked at a vendor. Now that I attend the occasional internal event at other companies as part of my job, I don’t tweet from those unless it’s explicitly requested, or unless what I’m saying is something the organizers are tweeting about themselves! The presence or absence of a hashtag is one clue, and I also have taken to just asking, “Do you want me to livetweet this?”

On negativity

It’s unlikely I’m going to overtly criticize you or tell you how wrong you are in the course of livetweeting your talk, since I think it’s really not so fun as a speaker to get off the stage and get hit by a wall of negativity in your mentions. (Not every person who livetweets feels this way.) I’ve noticed sometimes people use the conference hashtag and not the speaker’s Twitter handle, so you could try that if you’re into subtweeting in an obvious way. Not my style, as a general rule.

While I probably won’t tweet about this (because I’ve lived that movie already), if you’re telling us that your ops guys are going to configure things so a dev can do whatever he wants and product gets what he wants and apparently you have no women in your engineering org, I’m probably eyerolling about it in DMs or on Slack. And if there is something more than the everyday erasure of women, something so egregious I think the conference needs to take action, I would rather follow their posted code of conduct escalation process (or just DM the organizers if I know them). Public shaming isn’t my preferred place to start a conversation.

Arguing on the internets

“Yeah, I could do that, but I’m paralyzed with not caring very much.” Spike, BtVS 5x11 “Triangle”, 2001

From time to time, something I tweet strikes a nerve and suddenly lots of internet randoms jump into my mentions to “discuss” it with me and (sometimes) with the person I’m quoting. I’m almost never going to reply, because I’m not livetweeting the talk in order to engage in a spirited debate about the nuances of what the speaker actually might have meant (as experienced by someone whose only context is my tweet).

My mentions are busy enough during a conference that even if I want to chat with you on Twitter, it’s probably not going to happen. Sorry not sorry. I’m not ignoring you, but coming to say hi in person is probably going to work a lot better. I’m friendly, really! And although I tweet a lot during events, I’m mostly there to meet people in real time. We could all just watch conference videos on the internet if we didn’t want to interact.

On being wrong

I know you’ll all be shocked - shocked! to find that gambling is going on in here I’m a fallible human. This means that sometimes I don’t capture a speaker’s meaning quite accurately. I think it’s important to reply to my own flawed tweet with a clarification. (I’m not a huge fan of deleting tweets unless it’s right away for a typo fix, but I’ll reply to add context, especially when I try to paraphrase and miss the mark.)

Showing the audience

An engaged audience in rapt attention to a speaker is a nice thing to show, but I think about a few things - is the room too dark? are too many chairs in this part of the room empty, making it look emptier than it actually is? does every audience member in sight fall into the exact same demographics in a way that’s not representative of the conference? - before I decide if an audience reaction shot is going to be a net-positive. I’ve also taken pictures that I decide not to use due to content on people’s laptop screens being clearly visible (and I’m not going to sit there and blur it out on my phone - I’ll just take a refocused shot).

Being in multiple places at once

One of the great things about single-track conferences is not having to make hard choices. At a multi-track event, sometimes I want to tweet from more than one talk that’s happening in the same timeslot. I’ll tell the speakers ahead of time (if they’re co-workers or friends who’ll notice) that I’m not leaving or arriving mid-talk because I don’t care, but because I’m trying to be in two (or more!) places at once.

This is likely to happen if I’m acting as a track host at Velocity but also want to see a talk from another track. And sometimes I ask another person to livetweet a specific talk if I can’t be there at all! For example, my spouse Joe Laha did a great job with being voluntold to tweet from John Allspaw’s CraftConf 2016 talk while I was tweeting from Andrew Clay Shafer’s.

Adding to the joy in the world

This might sound like a lot of work, but it’s truly a lot of fun. I love connecting people and giving attention to thought-provoking ideas in tech, and I hear lots of positive reactions from people who get value out of my conference tweets.

As a speaker, I enjoy it a great deal when people tweet during my talks, and I even embed tweets in the talk page after the fact (like for my keynote at SpringOne Platform 2016). My greatest motivation to livetweet talks is that the speakers are putting work into giving me talks I’m learning from and enjoying; I’m happy to give them a 280-character review they can retweet, quote, embed, send to their parents, drop into their work Slack, or enjoy in whatever way makes them smile.

---

Thanks to Ryn Daniels and Tim Gross for beta-reading. All errors and #questionablelifechoices are my own.