Earlier quoted context omitted.
Is there a better way for governments to manage such processes ?
One of the major problems (at least in the US) is that the process is heavyweight and specifically tuned towards "lowest conforming bid". You end up getting subpar work and long delays, but the government is absolutely unwilling to restart the process in almost all cases because the cost is just that large (and we picked this one because it was going to be the lowest cost!). In some ways, the government has very litt…
IBM contract cost for failure-plagued Phoenix payroll system jumped to $185M
51–60 of 110 posts
Re: IBM contract cost for failure-plagued Phoenix payroll system jumped to $185M
#52Re: IBM contract cost for failure-plagued Phoenix payroll system jumped to $185M
#53Like all IBM projects now day the bulk of the work was done in India for likely less than $5 CAD/hr, while they billed out $100 CAD/hr or even more. The actual numbers may be slightly different, but still these are close to reality. What a great profit margin for a failed project. I wouldn't be surprised the code is like a lot I've seen from Indian-based sweat shops (btw there are amazing Indian coders, but you gener…
Re: IBM contract cost for failure-plagued Phoenix payroll system jumped to $185M
#54I saw failures on every level. The client (government), the PMs, the developers. It was mind-blowing how much incompetency was infested into the entire system.
I think the problem with the public sector is that some of them genuinely don't give a shit about their work and there's really zero repercussions to them for not doing their job well. They won't get fired like someone would in the private sector. With that said, I've no doubt some of them do work hard and are comparable to their private sector counterparts.
But until workers in the government can truly be held accountable, these types of IT disasters will continue to happen. Heads need to start rolling.
Re: IBM contract cost for failure-plagued Phoenix payroll system jumped to $185M
#55The public sector in a number of countries seems addicted to these large-scale high risk forms of procurement. Instead of running a larger number of smaller projects, which would be harder for them to adminster, there's a temptation to lump it all into one mega-project and shift the responsibility to a consulting company. The problem with that approach is, of course, that its almost impossible to accurately specify a…
Re: IBM contract cost for failure-plagued Phoenix payroll system jumped to $185M
#56The public sector in a number of countries seems addicted to these large-scale high risk forms of procurement. Instead of running a larger number of smaller projects, which would be harder for them to adminster, there's a temptation to lump it all into one mega-project and shift the responsibility to a consulting company. The problem with that approach is, of course, that its almost impossible to accurately specify a…
Re: IBM contract cost for failure-plagued Phoenix payroll system jumped to $185M
#57Re: IBM contract cost for failure-plagued Phoenix payroll system jumped to $185M
#58Earlier quoted context omitted.
The government creates a Request for Proposal (RFP) listing what the government wants. Companies have some listed period that could be as short as 7 days or as long as 6 months to respond with a Proposal of work including costs. Government then reviews and picks one based on the stated criteria. This might seem like a reasonable solution, but it adds significant overhead.
So adding onto this, if you do a quick google for an exmaple of a government RFP. OMFG! no agile in sight. You'll be greeted with about a thousand pages of spec. This is where the big shops earn their money. They know how to grind through the paperwork.
I am also aware of some huge pivots where some minor feature became the only thing the customer cared about.
Re: IBM contract cost for failure-plagued Phoenix payroll system jumped to $185M
#59Earlier quoted context omitted.
Is there a better way for governments to manage such processes ?
One of the major problems (at least in the US) is that the process is heavyweight and specifically tuned towards "lowest conforming bid". You end up getting subpar work and long delays, but the government is absolutely unwilling to restart the process in almost all cases because the cost is just that large (and we picked this one because it was going to be the lowest cost!). In some ways, the government has very litt…
It's stupid that for so long it's been coded into law that you need to spend as little as possible, because that's part of what got us here; the buyers in a lot of cases can't go with a better, more expensive option.
Hopefully the "best value" culture shift in government buying will go along with legal changes to let them act on that.
Re: IBM contract cost for failure-plagued Phoenix payroll system jumped to $185M
#60I once worked with a large team from IBM that was staffed with Indians that did the work both onsite and offshore in India and it was without a doubt the worst code quality I've ever seen in my career. There were classes with 10,000+ LOC, methods with over 1000 LOC and dead unreachable code that wasn't even commented out. Speaking of comments, if it wasn't automatically generated by the IDE there would be none. But the most horrific example of their lack of attention to detail and code quality was the absolute absence of any unit tests in this massive project with millions of LOC. The only unit tests ever written were by me and those unit tests were subsequently disabled/commented out when some other developer decided it would be easier to disable the unit tests rather than to modify them to adhere to the changes they were making. It's a shame more clients don't hire independent code auditing companies to inspect the code their paying millions of dollars for.