LogFAQs > #930424654

Then why does any chat conversation that escalates in complexity invariably result in one of us walking over to the other person's computer?

What I'm saying here is that if what you say is correct that chair swiveling is the most awful way to communicate and that the damage it has on my company is easy to demonstrate, the following scenario I'm about to describe would not exist. And yet it happens several times a day (not with just me specifically but across the whole department). We do not depend completely on face to face communication but it is far more practical in some contexts.

Here is an example scenario of what I'm describing:

I'm in the office, at my desk, working on a bug ticket. I'm looking through the relevant section of code and just getting a grasp of what it's trying to do. It's the first time I've seen this particular part of our code.

Eventually I pinpoint where things are going wrong, and I start trying to think of the best solution. A helpful thing to have would be full contextual knowledge of what the original developer was aiming for when they wrote the code - why do it this way and not that way, are there any pitfalls I should be aware of, any relevant business logic, do any of our other repos depend on this code, etc.

The original developer is sitting at their desk which is approximately 10 feet away. I look over and notice him sitting there with headphones on.

He seems to be focused, so I send him a message over our team chat as to not be intrusive (notifications are configurable, he can be bothered when he wants to be, etc. yay asynchronous communication)

After 5 or 10 minutes he responds something to the effect of "oh, hmmm that was a long time ago that I worked on that. I need a refresher to be able to give a good answer"

Now at this point, he has two options (well, three I guess):

-read over the bug ticket to get a gist of the issue, then change contexts of what he's working on, (opening up another instance of his IDE, browser, etc.), sift through the code long enough to jog his memory on everything, eventually type up some notes and/or response questions which may or may not ultimately be of any real relevance to my direct problem

or,

-he can stand up and take 5 steps over to my desk, where I'm sitting there with the exact spot in the code already open, and my mind is totally fresh on all the context of the bug and the existing code, so I can very quickly catch him up on the exact relevant parts, which things I've looked into already, what I've considered and not considered, etc. , and as he sort of stumbles around in his memory getting back to the point of the code path I'm working in, I can quickly steer him and keep him on track because I know exactly where I need him to be

(third option is he just doesn't help me but that's neither here nor there)

Now, that whole face to face interaction could take place over the team messaging system. But it would take way longer even if we said the exact same words back forth, because typing is slower than talking. But on top of that it's impossible that the exchange would not use more words over chat because I don't get to know what he's thinking until he fully completes and commits a message. I just see some moving ellipses for 30 seconds and then get a few lines that may not be remotely relevant. Whereas in spoken language I could cut him off right away if he's off base.

When given the option in this scenario, it's kind of silly to picture us sitting 10 feet part, chatting to each other about the problem, when we could just be talking and be way more effective. Everyone who chooses to work in the office would make the decision to stand up and walk over to the other person's desk, because it's just a thousand times more practical in that context.