I worked in corporate IT (at a huge bank) for 17 years, and --thank God-- just quit 6 months ago, EXACTLY for these kinds of power-plays by the IT-ignorant.
The worst part of these situations is that the whole process of centralized IT decisions is generally started by well-meaning managers who are appalled at anecdotal evidence of inefficencies, failures, and (in some cases) out-and-out corruption. Then the whole thing gets out of hand and turns into a "the geeks at IT don't know how to manage technology and align it with the organization's goals, so we have to tell them what to do at every step of the way."
Then the cycles of silver-bullet management fixes don't stop, and actually overlap: one "solution" hasn't even shown its possible benefits when the next one comes along. Management overhead then starts eating up more than half of IT's time and efforts, in a downward spiral.
The MA Commonwealth people have to see that micromanagement via committee (a.k.a. "task force") simply Does Not Work.
Committees are good for doing meta-management: setting agendas and guidelines, and using feedback to see if those guidelines are helping people follow the agendas. As part of that guideline-setting process, the committee MUST delegate responsibility, and permit ostensibly extra-agenda actions to be carried out, providing that the person who is doing that action has a way of proving that it really does follow the agenda, and it must permit them to prove it AFTER the fact. This is so people can do their job without waiting in line to get a bunch of signatures. If in fact the actions prove to be irrelevant to the agenda, then the bright IT guy will get sacked, but then the majority of IT guys who are following the agenda will be able to do the Right Thing without sitting on their backsides waiting for paperwork.
Well, in a nutshell, that's why I left corporate IT: I got tired of wasting time on committees and sitting on good ideas because too many people were covering their derrieres aginst the day that the centralized task force/committee/whatever will "(p)review" their projects. BTW, I was no bottom feeder in the IT department: I was two steps below the CIO/CTO, so I saw this up close and personal.
The problem is that spinning off a separate company ("again!") is like the sheepherd crying out "Wolf!" too many times.
Even if the product and company does last another ten years, what will happen if nobody takes it seriously again?
Our migration (or not) to Oracle is very dependent on what happens in the next six months. If things don't change, we'll have to pay a gazillion dollars to change over and ensure the continuity of our business (hint, hint for Borland).
We've been running our stock trading system with Interbase for over four years, with great results and performance: 150 *heavy* simultaneous users, over 8000 avg daily financial transactions, over 800,000 avg daily db transactions on a midrange HP-UX machine, avg response times for interactive operations are sub-second (simultaneous with other heavy batch stuff).
Unfortunately, we *might* have to move over to Oracle because we're still not convinced that Borland takes Interbase seriously, and most big-iron development tools vendors aren't convinced either: they're *not* updating their support for Interbase (most still support up to Interbase 4.0C).
Let's see what develops in the open source community. I'm hoping that much of the functionality lacking in Interbase that commercial DBs already have will be integrated by "open sourcerers", and then we might not have to use Oracle:-).
Slashdot Mirror
I worked in corporate IT (at a huge bank) for 17 years, and --thank God-- just quit 6 months ago, EXACTLY for these kinds of power-plays by the IT-ignorant.
The worst part of these situations is that the whole process of centralized IT decisions is generally started by well-meaning managers who are appalled at anecdotal evidence of inefficencies, failures, and (in some cases) out-and-out corruption. Then the whole thing gets out of hand and turns into a "the geeks at IT don't know how to manage technology and align it with the organization's goals, so we have to tell them what to do at every step of the way."
Then the cycles of silver-bullet management fixes don't stop, and actually overlap: one "solution" hasn't even shown its possible benefits when the next one comes along. Management overhead then starts eating up more than half of IT's time and efforts, in a downward spiral.
The MA Commonwealth people have to see that micromanagement via committee (a.k.a. "task force") simply Does Not Work.
Committees are good for doing meta-management: setting agendas and guidelines, and using feedback to see if those guidelines are helping people follow the agendas. As part of that guideline-setting process, the committee MUST delegate responsibility, and permit ostensibly extra-agenda actions to be carried out, providing that the person who is doing that action has a way of proving that it really does follow the agenda, and it must permit them to prove it AFTER the fact. This is so people can do their job without waiting in line to get a bunch of signatures. If in fact the actions prove to be irrelevant to the agenda, then the bright IT guy will get sacked, but then the majority of IT guys who are following the agenda will be able to do the Right Thing without sitting on their backsides waiting for paperwork.
Well, in a nutshell, that's why I left corporate IT: I got tired of wasting time on committees and sitting on good ideas because too many people were covering their derrieres aginst the day that the centralized task force/committee/whatever will "(p)review" their projects. BTW, I was no bottom feeder in the IT department: I was two steps below the CIO/CTO, so I saw this up close and personal.
The problem is that spinning off a separate company ("again!") is like the sheepherd crying out "Wolf!" too many times.
Even if the product and company does last another ten years, what will happen if nobody takes it seriously again?
Our migration (or not) to Oracle is very dependent on what happens in the next six months. If things don't change, we'll have to pay a gazillion dollars to change over and ensure the continuity of our business (hint, hint for Borland).
We've been running our stock trading system with Interbase for over four years, with great results and performance: 150 *heavy* simultaneous users, over 8000 avg daily financial transactions, over 800,000 avg daily db transactions on a midrange HP-UX machine, avg response times for interactive operations are sub-second (simultaneous with other heavy batch stuff).
:-).
Unfortunately, we *might* have to move over to Oracle because we're still not convinced that Borland takes Interbase seriously, and most big-iron development tools vendors aren't convinced either: they're *not* updating their support for Interbase (most still support up to Interbase 4.0C).
Let's see what develops in the open source community. I'm hoping that much of the functionality lacking in Interbase that commercial DBs already have will be integrated by "open sourcerers", and then we might not have to use Oracle