9
pleuph
7y

My team and I are working on a huge project that's been in development for years.

First deadline was in the fall last year. We were never going to make that.

Then we were supposed to be ready just after the summer holidays (months ago). We didn't make that either.

Then we were supposed to launch last week. Didn't happen, still too many critical errors and unfinished, untested features.

Now we are having daily meetings to discuss whether we'll be ready to release... that day!

Meanwhile, stability issues and other critical errors keep popping up. The product is barely finished and has not been through rigorous testing with all the latest features and bug fixes. Not to mention that we don't really have a deployment pipeline either.

And here's the kicker: The customers don't know this is coming. It's highly anticipated, but only internally. It is a replacement for an existing product, which strives towards not changing the frontend too much.

Why do we rush it so? I get that a deadline can help motivate you to reach your goal, but how motivated will we be if the launch fails and we get buried in bugs and missing features?
Would it not be better to launch it with at least the confidence of knowing that we've tried to test it properly?

Comments
  • 4
    This sounds like you could use a DevOps person/team.

    Basically it reads you lack:

    - Unit/integration/application testing.
    - ci/cd
    - config mgmt (puppet/ansible/chef...)
    - monitoring

    Meanwhile, stop going to those meetings and get things done and look at above items... Those meetings are a waste of time.
  • 1
    @NeatNerdPrime Yeah. All those tasks fall to whichever developer has the time. But to be fair, we are a very small team, and the company is not a software provider.

    We do have unit testing for large parts of the product, but it's not streamlined. The integration testing... yeah, that's the real problem.

    And we do have CI for our testing environment, but production is pretty much manual deployment right now.
    And we have discussed, but not implemented, branching.
  • 1
    @NeatNerdPrime Oh, and the meetings are like extended scrums, which do make sense as long as the plan is to launch ASAP.

    The problem is the hard push for launch.
  • 2
    @pleuph well, i would propose you would propose your company yo invest at least two months to get that pupeline streamed out, then you are able to get there without major hickups...

    It might seem long, but without it, your product will take many months/years longer to develop then it should.

    I'm always willing to help for a fair price , remote :)
  • 0
    @NeatNerdPrime Good idea. Won't work. I just don't have the gravitas to change people's minds at this time. We've all just accepted that the higher up's want this tested live.

    We have chosen the smallest of four areas to go live in first, and we will be able to roll back if necessary.

    But I do get your point. It's just so far out of my hands that it's not even funny.
  • 1
    @pleuph man that must suck donkeyballs... What if not just YOU but your entire team goes on strike?? That way management will have to listen if they want to get anything done...
  • 0
    @NeatNerdPrime Well. The thing is. We've all just sorta accepted that this is how it is.
    I have been too stressed out, personally, to shoulder the responsibility of speaking up.
    And almost half the dev team are still 'new', having been with the company for only half the development time.
    So there are many reasons why things are like they are.

    We are all still learning, and we do have great team spirit. So not everything sucks. I guess it's just this last big push that's making us feel really unprepared now.
  • 0
    @ClemFrieckie Could you rephrase that? I don't understand your question.
  • 0
    @ClemFrieckie Complete rewrite of a POS website with everything new from scratch, except for the underlying ERP.
Add Comment