The Effects of Technology on Work Productivity
This post is part of our Work Symposium. An introduction to the symposium can be found here; all of the posts written for the symposium can be found here
A professor once said that after optimizing his work processes in the twenty years after first using a computer he had finally become more productive with a computer than without one. He said that in the late 80s.
It was a laugh line, but the joke would only be funny if it were a bit true.
It’s worth remembering that technological products are developed largely to make money for their creators. If it can make you more productive along the way, the makers won’t stop that, but it might not be their foremost concern either. (See Candy Crush.)
Games are obviously unproductive, but not all productivity technologies are productive either. Some tools are useful for work communication but enable all sorts of other evils. (See e-mail.) Other tools provide important information but can also serve as a never-ending time sink (See Wikipedia.)
For the past two years, I have worked on a project whose groundwork was laid during a trip to China with a paper notebook and no computer. I don’t think I could have made such plans if I had access to a computer and its opportunities for productivish distraction.
In China, without a computer, when I had an idea, I would write it down and think about it. If I have an idea now, I research indefinitely on the Internet, and it moves no further.
Photo credit to Gary J. Wood. No modifications made.
The technologies I have access to determine what work I do. The work I do does not determine what products I use.
It really should be the opposite. I should have clear goals in my life and determine what tools I may or may not need to achieve those goals. Instead though, I find that I have acquired a bunch of tools for a lot of random reasons (it was on sale; Will recommended it). Those tools make certain things easy and fun, so those are the things I do. Goals, schmoals.
I am writing this post on my phone on an airplane. If I had free internet access, I might be reading news instead. If my phone weren’t available, I might be reading a book I packed that is important to my work. The technologies I have dictate what I do.
Steve Jobs would talk about delight in the use of Apple products. That is great if you are an Apple shareholder. If you are a customer, it can be a big problem. Are you using a tool because it is delightful or because it is the best way to get what you want done?
R is one of the least delightful statistical packages available, but it is much more important that I use it rather than the delightful and thoroughly unproductive Flipboard news app.
But maybe I shouldn’t even be using R. Maybe I should use a whiteboard to plan what analyses I need to do. Or I should use mind mapping software. Or I should read a book on a related topic.
Sometimes I wonder whether people keep checking their e-mail not because they can’t stamp down on their curiosity about who has just written to them, but rather because answering e-mail is something they know how to do well.
I am confident in writing e-mails. I don’t know whether I can do what I need to do in R. Part of being a researcher is doing things that haven’t been done before and you therefore don’t know will work. Maybe I would be better off if I didn’t have e-mail. I would be a less efficient communicator, but communication isn’t an important job responsibility of mine. Why did I decide to get e-mail in the first place?
Oh, right. I didn’t decide. E-mail is free and fun. Therefore I have it.
And I only thought to even ask what my most important job responsibilities are and what technologies assist or impede their fulfillment because I am on an airplane without internet but with a phone. When the Internet becomes truly ubiquitous and batteries last forever, maybe even these tiny opportunities for thinking about whether our technologies enable our work will be gone.
This is not to say that I am anti-technology. I am just observing that I have adopted a variety of technologies and associated behaviors without seeking even a nominal answer as to how they will help me do things that are important to my work. Rather, I use them because other people use them, I have made a habit of using them, and they are easy.
A tool used badly is not a failure of the tool.
I often see people say “oh, email makes communication worse, Powerpoint makes communication worse, Instant Messaging makes communication worse”. Not true. What’s happening is that bad communicators are now expected to do their own communicating, and we can all directly experience how bad they are at it, and their poor ability directly affects us.
It used to be that if someone were a bad communicator, they had a secretary (or a subordinate) to do the actual communicating for them. That person would type up the memos going out, and read the received memos coming in, and the superior’s inability to parse a sentence wouldn’t matter. There was a whole graphics department devoted to making presentations.
These days, you are your own secretary. You are your own graphics department. And the people who decided that secretaries and graphics departments weren’t needed (because we have email and desktop presentation-graphics packages) didn’t seem to realize that the secretary was more than just a human Dictaphone, that the graphics department was more than just a big printer that could do transparencies.Report
+1 to everything Jm Heffman says here, but I’d also add this:
Different technologies bring different things to the table.
For example, there’s no question that my using a laptop with internet connection provided distractions at my office over the years. But it also provided instant access to ideas, case studies, and people that I never had in the days where everything I did was either on paper or on a no-frills, DOS, closed proprietary system attached to my office’s desk top.
A rather large part of what allowed me the option to retire when I did was creating and implementing something no one in my industry had bothered to think much about. But that model wasn’t something that just hit me like an apple out of a tree one morning. It was the result of much exploring parts of my own industry — and a lot of other industries that had no initial connection to mine — that simply weren’t visible to me in my office, or even my entire company. Without a computer and internet connection, I would never have arrived where I did.
On the other hand, my model also wouldn’t have ever been realized if I didn’t take time twice a year to “sabbatical” for a week away from all people, TVs, commuters and phones, armed with only paper, pencil, and a few books.Report
When people ask me how I can spend a bunch of money for an Internet-connected smartphone, all I hear is, “How could you possibly waste a couple of Jacksons a month on a tool that gives you on-demand access to most of the collected knowledge of humanity?”Report
It pays for itself, in easily-answered and immediately-collected bar bets!Report
“…Instant Messaging makes communication worse”. Not true.
I largely agree with you, Jim.
I would say “instant messaging influences what kind of communication occur with which people at what frequency.” And it might determine whether you communicate at all. There are a lot of lame “Happy Birthday” messages out there thanks to Facebook. And as you say, it would be deceptive to say that “Facebook makes ‘Happy Birthday’ messages worse.” I would say instead that a lot of people who wouldn’t have otherwise paid attention to your birthday are now messaging you about it. And they are doing that instead of doing something else with their time because the technology makes it so easy.Report
What’s happening is that bad communicators are now expected to do their own communicating, and we can all directly experience how bad they are at it, and their poor ability directly affects us.
I mostly agree heartily with this, but I have a quibble with the first line.
A tool used badly is not a failure of the tool.
This part is kind of debatable… some technological tools are missing pretty basic features that would make them a lot easier to use while discouraging bad usage. Other tools have had features added to them that unfortunately make them ridiculously easy to use badly.
The reason why it’s only “kind of” debatable is that the root cause of those problems are usually that people use tools badly, though.
Take email. You’ve got two bad uses of email that illustrate both aspects of this.
Email was originally designed as person-to-person, terse asynchronous communication, on a single system no less. You logged into a system, you sent someone else on the system an email, and when they logged in they read it and they replied back to you when they got around to it.
Then systems started routing email between each other, which was fine and dandy back when setting up an email system was a task that required a bunch of access that required specialty knowledge and access, itself.
Then systems of systems began routing traffic between each other, which suddenly made the fine and dandy old system of routing email extremely problematic, because email systems are designed by default to trust each other, which is why we now have spam.
A technical decision (don’t validate senders) has led to a twenty year campaign against people whose adoption of email has entirely been malicious.
Also back in the day, there were no extensions to email. You could send ASCII characters, and that’s it. And then people thought, “You know, if I could send a *document* via email, that would be great”, and MIME was born.
And the birth was an abomination.
I’ll just quote the Wikipedia page on MIME:
“The presence of this header indicates the message is MIME-formatted. The value is typically “1.0” so this header appears as
MIME-Version: 1.0
According to MIME co-creator Nathaniel Borenstein, the intention was to allow MIME to change, to advance to version 2.0 and so forth, but this decision led to the opposite outcome, making it nearly impossible to create a new version of the standard.
“We did not adequately specify how to handle a future MIME version,” Borenstein said. “So if you write something that knows 1.0, what should you do if you encounter 2.0 or 1.1? I sort of thought it was obvious but it turned out everyone implemented that in different ways. And the result is that it would be just about impossible for the Internet to ever define a 2.0 or a 1.1.”
That, in and of itself, is terrible, but even the idea of “attaching documents” is antithetical to the original design parameters for sending mail: SMTP stands for “simple mail transfer protocol”; it doesn’t do any sort of sender validation (previously noted as a problem) but it also does no compression. So attaching a file to an email is the least efficient method of transferring that file to someone else via the Internet… FTP was a better alternative. Adding MIME to basic email would have been fine if mail was designed, top-to-bottom, as a loosely coupled system, where you could design an alternative mail transfer protocol… but it the real world it isn’t, because practically speaking nobody can use a mail transfer protocol other than SMTP, because all mail servers transfer mail using SMTP. The whole thing has become an ossified system.
The Network Effect is a harsh mistress.
So now you have people that use attached documents in email as a document version control system. There are people with tens of gigabytes of mail, largely due to N copies of M documents attached to P emails, for arbitrarily large N, M, P.
And nobody can throw anything away. And everybody expects email to be instantaneous, and synchronous.
There is a ton of blame to go around on why the beast is an ugly, ugly Chimera, but the tool designers deserve a chunk of it, too.Report
This was an awesome, if horrifying, post. Thank you.Report
Your comment reminded me of how email was addressed on UUCP a!b!c!d!user. You had to know the route to the users email server, which of course differed from system to system, the same person could have be e!f!c!d!user from another machine. Back then the email system was ascii only and you used uuencode and decode to send files at modem speeds. Just thought a trip down memory lane could not hurt.Report
The real problem is that having both the attachments and the main section of the email leads to Cartesian dualism (the MIME-body distinction).Report
These days, you are your own secretary. You are your own graphics department. And the people who decided that secretaries and graphics departments weren’t needed (because we have email and desktop presentation-graphics packages) didn’t seem to realize that the secretary was more than just a human Dictaphone, that the graphics department was more than just a big printer that could do transparencies.
It’s less the fact that the value of them are not understood*, it’s that “good enough” is frequently good enough. Much like how mass produced clothes aren’t as good as custom, hand-made clothes, but the trade-off in cost makes it worth it. Sure there is a trade-off due to lesser communications and worse design, but you have efficiencies due to the fact that you no longer as large an admin team or as big a graphics department.
* Witness that senior execs still have admins and agencies are still hired for the really important stuffReport
When I started at Bell Labs in 1978, there was a typing pool whose job was to turn engineers’ hand-written documents into type. Shortly thereafter, UNIX and its word-processing tools became generally available within the Labs. Over the course of 18 months the typing pool became a word-processing pool, and then disappeared and engineers had to do their own text entry. I was representative of the huge wave of young people the Labs had hired at that time and could compose at the keyboard. At one point I had to demonstrate to my department head that I was indeed composing that way, rather than wasting time by entering my hand-written draft myself. God only knows how many person-hours were lost because the Labs didn’t bother to make touch-typing classes available to the older technical staff. I remember watching older — that is, 40-something — engineers painfully hunt-and-pecking their documents.Report
For a long time, I didn’t understand text messages. We invented the telegraph. We then invented the telephone. We made the telephone wireless and computerized it to the point where it could send arbitrary digital data. And then we put a telegraph emulator on top of it.
And then one of my computer scientist friends pointed out that texting (and email, done correctly) are a great means of efficient non-blocking communication. Anybody who has written a complex piece of software with lots of I/O (networking, especially) appreciates how much easier it is to be efficient and responsive when you can start a transfer operation and then do other things while you wait for a response. Unless a lot of fiddly details need to be worked out, an interactive conversation when both of you have to be available and fully engaged at the same time tends to be fairly inefficient.
The problem arises when we get the priorities wrong. The fact that an email arrived doesn’t mean that you drop whatever you’re doing and handle it. You might have to, but it’s not necessarily the case. If you can handle your scheduling appropriately (and if senders understand that everybody should have a multiple-priority system for handling those asynchronous messages), you can avoid a lot of unnecessary communication and scheduling overhead if you go the non-blocking route.
A text message that says, “Call me,” isn’t stupid as long as you read it as, “Call me at your next efficient communications break,” rather than “Call me right now.” The fact that a lot of people don’t use it that way is the real problem.Report
It’s also true that if a person has trouble reading, then they are necessarily going to be slowed down by text communication. (See above about “maybe they’re just a bad communicator”).
I’ve worked with people who absolutely could not engage with text communication. You could go sit in their cube and solve the problem in five minutes face-to-face, maybe ten minutes over the phone, but if you sent them an email they’d take fifteen minutes to read it–not even ignoring or slow-rolling, it really would take them that long!–and then another thirty to reply.Report
Yeah, boy howdy on that one.Report
The terms I usually hear when people talk about that are asynchronous and synchronous communications. It also took me a while to figure out why texting was so popular. In retrospect, I felt stupid for not having realized. Calling someone is relatively “heavy” in its demand on the receiver. You are asking for their attention now. Most of the texts I send now are FYIs.Report
Exactly. It’s interesting that humans have naturally settled into this pretty well without even realizing it, because it has been known for a long time in the computing world. For example, the bad old days, the processor might ask the floppy drive for some bytes. It would then sit there and wait for the bytes to appear. The problem is that the CPU is potentially hundreds or thousands or even millions of times faster than the device it’s talking to, so every microsecond spent talking to the device is a microsecond wasted. It would be like you putting your clothes in the dryer and then sitting there until the dryer finished the job instead of going back to work and waiting for the buzzer.
There’s a lot of stuff that needs tight-loop back and forth. My team is completely distributed and uses email for most things, but we have a weekly standing phone call and occasional real-time chat or phone calls for the fiddly problems.
Email has the added advantage of giving the person you’re talking to time to think about or research your questions. Nothing irritates me more than going to a meeting and having a question sprung on me that I could easily have researched beforehand and answered in the meeting if they had simply emailed it to me. “Frog, what will the processor for new Machine that Goes Bing cost?” “I don’t know. If you had emailed me the question yesterday when you scheduled the meeting, we’d be using this meeting for me to communicate the answer instead of using it to ask the question and scheduling another meeting for the response.”Report
So, a bit of a rant on communications….
My company recently installed some form of instant messenger on our computers. Now, in addition to having phone, email, person to person, I have this IM function. So I read an email from a PM, then get a IM from her asking if the task is done. Really, how many forms of communication do I need?
I’ve sat in my office and heard my co worker call my boss, on speaker phone, who answered, on speaker phone. They reside less than 20 feet from each other. I heard each side of the converstion twice. Jeebus frickin christ.
In line with the above rant, people will not get up off their ass and go talk to someone. They call, or IM, or email. They are “doing something”. What they are really doing is wating for a response. Activity is not productivity! You’re just moving the balls around on the floor. Go make something happen and get something resolved. OMG.
Yeah, these aren’t 20 year olds either….Report
I find that there are people who will tend toward idleness given any excuse, and communications “hot potato” is one of the best. As long as they can respond to a request to do something with an inane question, they can avoid doing anything until that question is answered. Somebody skilled in this art can be “waiting” on everybody at any given time and playing solitaire 95% of the time.Report
Often one of the purposed of IM or email is to say you passed on as information (as little as possible) or made contact without actually having to make contact that solves the issue. It is like leaving a phone message for someone after their office has closed. It’s more like a game of tag….”tag i called you, now you have to call me to actually make contact”Report
@damon
IM is indispensable if you work with people who are out of your line of sight, especially if it has a function that tells you if people are at their computers. Waiting until someone is free to go talk to them makes life a whole lot easier.Report
@james-k
Yes, I agree it can be usefull, but see my comments about misusing tech above. And I still maintain that if you want to actually get something done, get a decision made, you go personally to them face to face and get a decision. Standing in someone’s office door usually gets an much quicker response than email, phone, im. The key thing is to no over use that. If I show at someone’s door I need an answer soon. 99% of the time, I’m supporting the person who’s door I’m standing in so it’s in their interest to answer my questions and let me get back to doing work for them.Report
Really, how many forms of communication do I need?
You “need” one form.
That form of communication needs to handle different modes, though:
Synchronous bidirectional verifiable: you and I need to communicate, directly, right now, back and forth, in order for the communication to produce effective results. Example: we need to figure out how to build this device to fit in this box using only these parts in the next half-hour or the astronauts on Apollo 13 kick the bucket. Implementations: phone teleconference, meeting.
Asynchronous bidirectional verifiable: you and I need to communicate, directly, back and forth, but without time pressure, in order for the communication to produce effective results: Example: I need my contractor to know what style of siding I want to put on my new house, but the foundation isn’t poured yet and the siding won’t need to go up for quite some time. Implementations: email.
Immediate Broadcast, verifiable: I need to let a bunch of people know a common goal, right now, and they must get the message. Example: fix bayonets, and charge! Implementations: bullhorn, instant messaging.
Immediate Broadcast, non-verifiable: I need to let a bunch of people know a common resource, but I don’t really care if they all get the message. Example: there’s leftover food in the break room, it’s going to get tossed at the end of the day, but if you haven’t had lunch yet, free grub! Implementations: bulletin board, forum.
The thing is, you don’t really need all of those various implementations, you need one that does all that stuff.
Nobody’s built it yet, though.Report
“riding in2 valE of death, brb”Report
lolwutReport
charge of the leet brigadeReport
R’s reputation is somewhat undeserved, after all even Excel is only user-friendly if you’re trying to do one of the tiny range of things Microsoft thinks should be easy to do. If you want to do anything complicated in Excel you need to code VBA.
Serious question, what would you be doing in R that you would be able to do on a whiteboard?Report
Oh, I wouldn’t be doing the same thing on R or on the whiteboard. But I will often plan out what sequence of analyses I want to perform in R on the whiteboard first. Then I might dig into R and get an unexpected result. At that point, I need to either go back to the whiteboard and revisit or push further in R. I’m not confident that I make the decision between which of those two things to do rationally rather than by habit and convenience. (My whiteboard is in another room and may require erasing.)Report