Stories
Slash Boxes
Comments

SoylentNews is people

posted by janrinok on Monday October 10 2016, @04:37PM   Printer-friendly
from the if-only... dept.

There is an interview with Joel Spolsky on GeekWire which reports that companies should Just shut up and let your devs concentrate:

If you want to attract and keep developers, don't emphasize ping-pong tables, lounges, fire pits and chocolate fountains. Give them private offices or let them work from home, because uninterrupted time to concentrate is the most important and scarcest commodity.

That's the view of Joel Spolsky, CEO of Stack Overflow, a popular Q&A site for programmers, who spoke this morning at the GeekWire Summit in Seattle.

"Facebook's campus in Silicon Valley is an 8-acre open room, and Facebook was very pleased with itself for building what it thought was this amazing place for developers," Spolsky said in an interview with GeekWire co-founder Todd Bishop. "But developers don't want to overhear conversations. That's ideal for a trading floor, but developers need to concentrate, to go to a chatroom and ask questions and get the answers later. Facebook is paying 40-50 percent more than other places, which is usually a sign developers don't want to work there."

[Continues...]

Spolsky, who in 2011 created project-management software Trello, said the "Joel Test" that he created 16 years ago is still a valid way for developers to evaluate prospective employers. It's a list of 12 yes-no questions, with one point given for every "yes" answer:

  1. Do you use source control?
  2. Can you make a build in one step?
  3. Do you make daily builds?
  4. Do you have a bug database?
  5. Do you fix bugs before writing new code?
  6. Do you have an up-to-date schedule?
  7. Do you have a spec?
  8. Do programmers have quiet working conditions?
  9. Do you use the best tools money can buy?
  10. Do you have testers?
  11. Do new candidates write code during their interview?
  12. Do you do hallway usability testing?

"The truth is that most software organizations are running with a score of 2 or 3, and they need serious help, because companies like Microsoft run at 12 full-time," Spolsky said when he created the test. He said that remains true today.

How well does your organization support its developers? If new or better equipment would improve your productivity, is it made available to you? How is your work environment? How well does your organization score on the 12-point "Joel Test"? What is the biggest thing blocking your company from improving?


Original Submission

 
This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 2) by JoeMerchant on Monday October 10 2016, @07:24PM

    by JoeMerchant (3937) on Monday October 10 2016, @07:24PM (#412590)

    But, scientists are precious flowers that can't be bothered to learn things like "real coding practices."

    I was hired by a special flower a few years back, he gave us his Matlab prototype code to "convert to a pretty GUI and clean up to make it run a bit faster." He was visibly disappointed when we found and implemented a 100x speedup in the logic that meant he wouldn't be needing the rack of supercomputer power and instead could do the job in real-time on a laptop.

    The highly advanced software development practice of a (singluar) code review led to that little breakthrough. It also crushed the ego of a fragile PhD and ultimately sent him crying back to mommy. We were as supportive and non-judgemental and kind as humanly possible, but some 35 year olds still haven't matured enough to survive having a mistake pointed out. Most of the people I've met who have these problems also hold PhDs. (Equal time to great PhDs - there are a ton of them out there, but just holding the degree is not a guarantee of greatness, or even adequacy.)

    --
    🌻🌻 [google.com]
    Starting Score:    1  point
    Karma-Bonus Modifier   +1  

    Total Score:   2  
  • (Score: 2) by shrewdsheep on Monday October 10 2016, @08:49PM

    by shrewdsheep (5215) on Monday October 10 2016, @08:49PM (#412629)

    While I don't argue your general sentiment, I would like to add that your experience is a matter of disjoint expertise. Most scientist do not have training in programming and are rather self-educated. That of course leads to the one-eyed in the land of the blind phenomenon. Yes, scientific code should have a score of at least about 4. On the other hand, do not optimize prematurely. The supercomputers are there, so by all means run your model there. Computer time is cheap as compared to the researcher's time. If you need to run it on the laptop, leave it to the experts. And please, package your software, I do not want to try your makefile that has no chance of running elsewhere but your own computer. In this sense, besides attitudes maybe, your story to me sounds as if things worked out rather well.
     

    • (Score: 0) by Anonymous Coward on Tuesday October 11 2016, @02:08AM

      by Anonymous Coward on Tuesday October 11 2016, @02:08AM (#412756)

      The book smart PhD dude was totally crushed by the street smart coder. This should be made into a movie or a cartoon or something.