[agentshark]

Hello fellow devs. I want your opinion about this job I ran into. I want to know if you would take the job or walk away after reading about its description.

Job
===

• A large half finished php project, estimated over 60,000 lines of php, html and css.
  
• Zero code documentation. Not even a basic entity relationship diagram.
  
• Zero unit tests. See last point.
  
• Developed in CakePhp 1.3 (the one compatible with php4) with a lot of code duplication, due to some validation being in controllers and not in models.
  
• Uses subversion where all developers push and pull from trunk only.
  
• Code has instances of raw sql injection prone code.
  
• Managers do not want to change how anything is done, nor implement anything new. Just finish by this certain release date.

[Da:Sourcerer]

Well, how much do they pay? It certainly doesn#t sound like something I'd put in my CV.

[billgudule]

No way ! I would run away as fast as possible !

[jacmoe]

I would take the job.
Because I am confident that I am able to tackle it, and it's a useful experience.
Personally, I don't care about unit tests, and documentation.. I can generate a diagram from the database - it's MVC, isn't it?
Even if it's CakePHP. 1.3 is not too bad.
I would take it.
Because there are lots of projects like that out there in the real world.

[agentshark]

I thought I could of generated a diagram also, but the tables are myisam not innodb. No foreign keys exists at the db level.

• The whole team also does want to change, not just the managers.
  
• Project going for over a year. Deadline in a couple weeks.

[Da:Sourcerer]

agentshark, on 11 May 2012 - 08:15 AM, said:

myisam

And ... it's a large project?

[agentshark]

45 controllers. Is this large?

The whole business IS the website. Many complex interactions and features. Not just a regular push out information type site.

[jacmoe]

If the deadline is in a couple of weeks and there is a team working on it already, I'd say: screw them.
I would take on a project like that if the deadline was a couple of months - depends on the existing team. But not in this situation.

[SapporoGuy]

If the pays is overwhelmingly good, yes!
If not, I'd say, "I want more money!" if they say no, just say you need it to compensate for battle scars you'll get.

Feel sorry for the managers too. They're on a budget and timeframe.

[rapidtransit440]

If you want to take the job, ideally explaining to the managers that their target date is unrealistic, because there current codebase was developed by someone else that didn't follow best practices, you can't guarantee your work because you can not depend on someone else got it right.

I can draw parallels with people in the plumbing and electrical trade, that got into similar situations and walk away.

[harmar]

Well depends on how much they pay, but I don't think the stress levels would be worth it, specially if the managers are dead set on how things should be done