Today wasn’t a sad day; I’ve been trying to offload the Client development/maintenance work for about 2 years now. But it has taken some final pushing to get rid at the end.
Today, I have logged jobs in our helpdesk to get the final batch of windows 2000 clients rebuilt across from my client base to the corporate (not mine) XP client, this all came about because a couple of years ago, some what I consider purely political decisions where made not to go with the client I had developed but instead start again (the fact that they reinvented large portions of the wheel doing this, which seems strange when that was in-fact what I was accused of).
One of the more disheartening things you can do in work is support a contemporary system, which has no future. So I have understandably being trying to get rid for a while, and for some reason people seem to think I have been holding back/blocking the transition.
Anyway it’s going now. Which is sad in a way, because I still believe as a system it had some neat ideas in it, I mean I long ago came to the conclusion that no matter how many neat underlying ideas on how to get an operating system on to a PC you can’t protect it from badly installed and configured software.
Also I think as an institution we will lose out, because not only have these ideas being lost, but we will lose the potential for more, because through the process of empire protection that started this all, we have lost cooperation that gave us the ability to improve and develop some quite cool client software. It’s not that people won’t work together (although that is also true), or that the people who would develop these ideas have been trodden on to much (although again this is also true), it’s that the development of clients has become extremely insular, and impenetrable as a process to all but a few, and the script kiddie ideology has become the one we now go with. Not that scripting doesn’t provide many great solutions, it’s just when we hit a problem with scripting, we work out the most complicated solutions to solve the problem, instead of doing what we where once good at, and blasting right through the problem with the talent we have in the department.
One example of this is a thing called the software launcher, early on in the history of clients in the university, we realised that we had far to many pieces of software to install all at once on the clients (around 400) so the solution was to code our way out of the problem with what in it’s currently carnation is something very akin to Install on demand, where the users run the software and it is installed as and when it is needed. As a result of this software our client systems where considered to be leaps and bounds ahead of other universities at the time (now with the rise of Windows Installer, effectively this system is redundant, but this was years before Windows Installer).
If at the time we first got this problem the attitude that is dominant now was in place we would have never gotten to such a neat solution, I would hazard a guess at an extremely complicated system of scripts elevate privileges and hacking files together, requiring quite a lot of babysitting and crossed fingers, nothing above adequate, and that’s the sad bit. Once as a department the client was considered our flagship product, and we took pride in the innovative way we solved problems to produce good clean, and easy solutions, which set us apart and hopefully above other university IT systems. Now we strive for an adequate, run-of-the-mill nothing special, system that makes us sink into the background of computing departments nationwide, while internally people are discouraged from innovation.
Gosh I love my job.