Computer Programmers and "Black Magic"

This week the 810 readings were all written by Nathan Ensmenger and all about the history of computer programmers from the 1940s to present.

I hadn't realized that the earliest forms of programming created a separation between a planner and coder, and that the coding aspect was considered menial labor. Ensmenger says that the transistion from coder to programmer was partially the result of the coders needing to rework the planner's work in order to make computer software run.

He uses this theme as a launching pad to discuss the idea of programmers as practitioners of 'black magic.' According to Ensmenger programmers were seen as highly creative individuals (at least in the late 50s and early 60s) who were irreplaceable. They practiced a black magic by doing a job that no one else could do. Cool shit.

Later, as the shortage of decent programmers was recognized, managers tried to 'dumb down' the profession and create ways in which they could train anyone to program (easier computer languages, training sessions, etc.), essentially taking the 'black magic' out of the profession. Ensmenger states that this tactic didn't work, and that the organizations that did this ended up with a lot of sub par programmers.

Which leads into the next major motif of Ensmenger's papers. What makes a good programmer? Apparently, extensive psychological testing and recruitment efforts have been made by industry to obtain good programmers in the past. Again, according to Ensmenger this has had less than efficient results. Although I haven't read all of his (Ensmenger's) work, it appears that there was never really a good solution for hiring or recruitment of programmers.

Finally, in an effort I'm leery of, Ensmenger tries to create a niche for technicians and programmers existing in limbo between white and blue collar workers. He advocates looking at these professionals with a different lens than other workers in order to provide another lens for industry in general. In order for me to buy this, I need more reasons why programmers are more individualistic than any other niche of worker, what exactly qualifies as a programmer, and how we are to look at this programmer niche historically.

All good readings, and very interesting. The 'black magic' bit and psychological testing aspects of the readings made them fun to read.

Uncovering Labour in Information Revolutions, 1750-2000

The readings for class this week covered less abstract topics and looked at four case studies about "information labor." The following is a short recap.

Hands and Minds: Clerical Work in the First ‘Information Society’ Eve Rosenhaft
German clerical workers from the 19th century go bonkers when their workload skyrockets. Divorce, mental illness, and arthritis make this an interesting read.

Compressing Time and Constraining Space: The Contradictory Effects of ICT and Containerisation on International Shipping Labour Helen Sampson and Bin Wu
Sailors go bonkers when ICTs alter industrial shipping by decreasing shore leave and providing a window an outside world that they can't take part in.

‘Computers in the Wild’: Guilds and Next Generation Unionism in the Information Revolution Chris Benner
Chris Benner compares Silicon valley 'guilds' to medieval guilds, professional associations, and unionism and attempts to show that the new guilds are a result of an increasingly volatile work environment in a network world.

Emerging Sources of Labor on the Internet: The Case of America Online Volunteers Hector Postigo
AOL volunteers go postal when the company tries to reorganize its business. AOL takes away many of the benefits that it gives volunteers and expects things to just be 'OK' (like the soda of yore). Lawsuits ensue.

There was also an introduction by Aad Block and a conclusion by former information laborer (and instructor of LIS 810) Greg Downey which provided a theoretical framework and some suggestions for future research. My main problem with the reading for this week is the ambiguous definition of information labor. In all the cases presented we see hidden labor problems, but why is this labor necessarily information labor? The idea of information labor is so general case that it could be taken to mean anything that involves human communication (information). I feel like someone hung up a piece of Velcro, labeled it as information, and started tossing garbage at it; Anything that sticks qualifies as information labor.

Other than that minor complaint, the material was a fun read. Maybe it's the nerd in me, but I liked the comparison of the new tech guilds with old medieval craft guilds. The other readings were equally nerd friendly, and should make for interesting discussion in class.

Reading Stalder's Castells

After gritting my teeth through the first five chapters of the book, it became clear that chapter six is the payoff chapter. Chapters one through five read like a gigantic catalog and present Castells's ideas in autonomous modules. After the short background section it seemed like Stalder just started naming off ideas that Castell's has thought up, and that he was leaving out the connecting themes.

In essence, he describes a compendium of 'nodes.' We start off reading about what Castell's believes is happening in distinct social spheres, move on to his ideas about social movements, and then later to his thoughts of space and time. In the way that the 'nodes' are presented, they come off as distinct identities that aren't described as interactive, and we are left with a handful of chess pieces with no rules or playing space.

So in retrospect, chapter six is the 'aha' chapter. Stalder finally tells us how to play the game with the pieces that he has given us, and presents Castells's major theory of networks. In a way, the arrangement of the text helps to understand Castells's network theory. Because the theory depends on modules that may drop in or out of the theory at any time, yet depends on each module for its overall presence at any moment in time, the presentation of the nodes before the connecting theme was a useful way to write the book.

At the same time, I feel like this is a weakness in the theory. Stalder also speaks about this in the book. If a phenomena cannot be adequately explained by Castell's Network Theory, it can too easily be explained away as another autonomous 'node.' Still, I have to throw out mad props to a guy (Castells) who essentially says (pg. 34-38) if my theory don't work for you, don't use it: It's only as good as the insight that it provides to the user.