Go, Suity!

[suityou01]

After all the tulipe you waffle about doing a good job and going through proper process it is clear you are technically incompetent and not able to deal with different working environments, what contracting is essentially about.

Why don't you go back to churning out hacked ASP in a basement for some third rate software supplier, that is all you are good for.

[MarillionFan]

Suity done a whoopsie

That update was a joke? Right?

[Old Greg]

That update was a joke? Right?

It's beautifully crafted if it is.

[suityou01]

That update was a joke? Right?

Darmie was too thick to understand my update with the lots of words, so I broke it down a bit for him.

[MarillionFan]

It's beautifully crafted if it is.

It has to be a joke post otherwise Suity is really Mr Bean.

[suityou01]

It has to be a joke post otherwise Suity is really Mr Bean.

Oh you mean the big post Darmy couldn't get?

Nope, that's real.

[suityou01]

Serious question. I can fix this. It would mean going off plan, doing what it takes, and getting ClientCo to give the system an end to end test in live (really not very professional), but it would get them up and running again.

Cheap and dirty fix, with all the same problems the next time we do a new release.

So quick and dirty, or stand by your guns and do it right?

[Ticktock]

I'm not techie enough to know the risks of your proposal.

In my world however, I'm always a fan of doing what is required to get the job done, and worry about the process later. Then again, generally there is very little that can cause problems that can't be fixed in what I do.

[suityou01]

I'm not techie enough to know the risks of your proposal.

In my world however, I'm always a fan of doing what is required to get the job done, and worry about the process later. Then again, generally there is very little that can cause problems that can't be fixed in what I do.

Risks are high. Bit of background. The system is due to be replaced so they don't maintain it, don't want to spend money on it. These fixes are only being done as they are critical.

Releases can only be done after 5pm. No one is prepared to work past 5 to stay and test. No test environment so it's pot luck when promoting from dev to live.

The only way to do this quickly is gangnam style. But if that goes wrong, I have no plan to refer back to and it's all on me.

Doing it right means patching up the environment (patches aren't cumulative) so it's a lengthy process and would need a full regression test which is again lengthy.

I inherited a lot of technical debt, and have no one on my side. The future's bleak, the future's brown.

[MarillionFan]

Serious question. I can fix this. It would mean going off plan, doing what it takes, and getting ClientCo to give the system an end to end test in live (really not very professional), but it would get them up and running again.

Cheap and dirty fix, with all the same problems the next time we do a new release.

So quick and dirty, or stand by your guns and do it right?

Groan I'll bite.

You should be pushing for a proper deployment process. The client is adamant they don't need this. You need to provide a risk analysis / associated costs versus possible downside of not doing it.

Every time you have an issue on dev to production you need a RCA. On that basis you need to be showing the associated risks / costs. Everytime you go to deploy I would raise a risk and then again raise each fail.

Short cuts = risk. Make the stakeholder / owner make that decision after presenting the evidence .