5 ms·

"I spent a day researching" - when software engineers are given reasonable amounts of time, great results can occur.

by bitops 14y ago

Rudism 14y ago
At my last job, I was always boggled by the architecture team, who would lock themselves in a meeting for 8 hours and completely design some complex system with just their notepads and a whiteboard. I used to think they must be über geniuses who already know everything needed to come up with the optimal design and algorithms without needing to do any research. In hindsight, with a few more years under my belt, I now realize that what I mistook for knowledge and skill was mostly just hubris on the part of most of the team members (which also manifested itself in strong objections to any outside suggestion that there may be a better way to do things). Challenging yourself to come up with better ways to do things is a great way to keep you learning, keep your work interesting, and improve the quality of your projects.
- ComputerGuru 14y ago
  At my old job, the other architects and I would spend days researching on our own, then lock ourselves up for days in a glassed meeting room with nothing but a notepad and a whiteboard and lots of coffee as we debated all the various methods we'd discovered, pointed out the flaws in each others' approaches, picking one person's research of choice and seeing how far we could push it and when/where/how it would fail in an attempt to reach the best solution.... then do more research, get some developers working on prototypes, and repeat the whole thing all over again in two weeks' time. All I'm saying is, it's not necessarily hubris. Too often you find half-baked or half-researched ideas that work in theory but not in the real world or the opposite (being things that work great but in practice could only ever be used in that particular application). You need to look into what transpired before and after those 8 hours meetings before coming to that conclusion (not saying you haven't).
  - gruseom 14y ago
    Why wouldn't the architects work on prototypes? In my experience, class distinction between "architects" and "developers" is a red flag. Actually, the term "architect" is itself a red flag. Come to think of it, even the term "developer" is kind of a red flag. A lot of red flags :)
    - ComputerGuru 14y ago
      Oh they did. I was a developer/architect, did more developing than any of the developers and more architecting than any of the architects. However, not all the "architects" (mind you, none of us actually used that term/title) were developers. We had the head of the QA in charge of testing and real world deployments on the team, as well as the lead interface designer. Their feedback helped to find a solution that would be both feasible and user friendly. That's why I said the developers would work on the prototypes.
      - gruseom 14y ago
        mind you, none of us actually used that term/title What terms/titles did you actually use? By the way, I didn't intend my "red flag" comment to imply that your particular setup didn't work well. There are lots of good variations. The sociology of software projects is fascinating.
        ComputerGuru 14y ago
        Team Leaders. Even if a team was more smaller teams :) C/C++ Team Lead, Web (C#/ASP) Team Lead, QA Team Lead, Design Team Lead, and The Boss (startup CEO).

"I spent a day researching" - when software engineers are given reasonable amounts of time, great results can occur.

by bitops 14y ago

"I spent a day researching" - when software engineers are given reasonable amounts of time, great results can occur.

At my last job, I was always boggled by the architecture team, who would lock themselves in a meeting for 8 hours and completely design some complex system with just their notepads and a whiteboard. I used to think they must be über geniuses who already know everything needed to come up with the optimal design and algorithms without needing to do any research. In hindsight, with a few more years under my belt, I now realize that what I mistook for knowledge and skill was mostly just hubris on the part of most of the team members (which also manifested itself in strong objections to any outside suggestion that there may be a better way to do things). Challenging yourself to come up with better ways to do things is a great way to keep you learning, keep your work interesting, and improve the quality of your projects.

ComputerGuru 14y ago
At my old job, the other architects and I would spend days researching on our own, then lock ourselves up for days in a glassed meeting room with nothing but a notepad and a whiteboard and lots of coffee as we debated all the various methods we'd discovered, pointed out the flaws in each others' approaches, picking one person's research of choice and seeing how far we could push it and when/where/how it would fail in an attempt to reach the best solution.... then do more research, get some developers working on prototypes, and repeat the whole thing all over again in two weeks' time. All I'm saying is, it's not necessarily hubris. Too often you find half-baked or half-researched ideas that work in theory but not in the real world or the opposite (being things that work great but in practice could only ever be used in that particular application). You need to look into what transpired before and after those 8 hours meetings before coming to that conclusion (not saying you haven't).
- gruseom 14y ago
  Why wouldn't the architects work on prototypes? In my experience, class distinction between "architects" and "developers" is a red flag. Actually, the term "architect" is itself a red flag. Come to think of it, even the term "developer" is kind of a red flag. A lot of red flags :)
  - ComputerGuru 14y ago
    Oh they did. I was a developer/architect, did more developing than any of the developers and more architecting than any of the architects. However, not all the "architects" (mind you, none of us actually used that term/title) were developers. We had the head of the QA in charge of testing and real world deployments on the team, as well as the lead interface designer. Their feedback helped to find a solution that would be both feasible and user friendly. That's why I said the developers would work on the prototypes.
    - gruseom 14y ago
      mind you, none of us actually used that term/title What terms/titles did you actually use? By the way, I didn't intend my "red flag" comment to imply that your particular setup didn't work well. There are lots of good variations. The sociology of software projects is fascinating.
      - ComputerGuru 14y ago
        Team Leaders. Even if a team was more smaller teams :) C/C++ Team Lead, Web (C#/ASP) Team Lead, QA Team Lead, Design Team Lead, and The Boss (startup CEO).