Subj : Re: Personal connections To : comp.programming From : rem642b Date : Mon Sep 05 2005 08:18 pm > From: Patricia Shanahan > I think this embodies a basic misconception. A programming team can > be much more effective than its members would be working in isolation. I think you're arguing a false dichotomy. In my major job (CAI-Calculus), we brainstormed general ideas, and debated various proposals, and sometimes debated details of interfaces, in a group. But each of us had different programming tasks to do so we mostly wrote actual code without disturbances, then punched sources files to each other for alpha testing. (VM/CMS had a "virtual deck of cards" mechanism for sending files between accounts on a single machine, more convenient and reliable than trying to include a program source within an e-mail.) > People tend to communicate better with people they know than with total > strangers. Socializing facilitates exchanges of information and mutual > aid. The benefit in "whole greater than sum of the parts" effects of > getting to know colleagues often outweighs the costs. None of my previous employers saw any benefit to such socializing. At no time did my supervisor or the head honcho or anyone else in a position of authority ever facilitite any of us meeting each other except as absolutely necessary for collaborating in work tasks. > I've known managers discussing proposed office space layouts to treat > spaces that support informal, casual conversation as an important design > feature. We were basically forbidden to talk with others when they were trying to work. In particular, at one point I was asked to teach myself enough to be able to write font-handling software in X, the window system on Unix. All I had were the huge volumes of X manuals. I found the relevant section, and wrote a simple test program as a first step toward displaying text on an X window. The program artificially generated a 1-by-8 bitmap and then tried to display it on-screen. It didn't work. I couldn't find anything in the manual that explained why it didn't work. I tried to ask help from somebody who already know how to do some X window-system programming, but I was ordered *not* to ask anyone for help even the most trivial beginner help as I needed. If I couldn't figure it out all by myself, then I should just give up. We weren't allowed to install any newsgroup software, and I didn't have my own ISP dialup account, so I couldn't ask anyone on a newsgroup. So after much frustration I finally had to just give up. To this day I still have never found anyone willing to tell me what I was doing wrong. (After 1991 when the project ended and nearly everyone got laid off due to lack of money for any followup project, I lost access to the Unix workstations, so when I finally got newsgroup access in 1992 it wouldn't have done any practical good to post a question, so I haven't bothered to post a vague question because I no longer have access to the test rig I was trying.) > Of course signals, such as closed office doors, are needed to get peace > and quiet when needed, but that should not be all the time. The only time I ever closed my office door was when I was taking a nap on the floor of my office, and when somebody was smoking in another office or somewhere else in the building or right outside the open front door with the stench of the smoke blowing into my office. My office was at the end of the hall, so I could see people walking along the hall, but there was maybe only once or twice in all those years that anybody stopped to say 'hi' to me. I have no idea what it's like to work in a place like where you apparently work, the kind of workplace you describe. .