It's all very well for Tea to fiddle about telling you things that you already know, but what you
really need to think about is the things that you
don't know. Viz:
- What is the basis on which your budget was set?
- Who makes the budgetary decisions?
- Are they reasonably well aware of the costs and benefits of different strategies when it comes to computer purchase?
Your job, as I see it, is to translate the technical realities of the situation in such a way that they, the decision makers, can clearly see what the choices are, and can then make an appropriate purchasing decision in possesion of all the facts.
I suspect that, at present, they are operating on the time-honoured management philosophy that you work out the absolute minimum cost to get whatever needs to be done taken care of, and then reduce it by an arbitrary amount, typically 20%. Having done that, you then hand the entire responsibility for the purchase over to an underling As soon as anything goes wrong, you fire the underling, and hire one who will work even longer hours for less money, and tell him to do the same job on 25% less than is actually needed. (Alternatively, you can
not fire the underling, simply haul him over the carpet for his gross errors and extract yet more sweat from him in future, not to mention his gratitude for not sacking him in the first place.)
This is how an alarming number of small businesses run. (I could give you a list of Ballarat and Melbourne businesses that work exactly this way, starting with the local technical training college, who are notorious for it.)
Or, of course, it could be that they simply don't know what computer equipment costs, or that they simply don't have any more money in the kitty and the $25,000 is all that is availabe.
So, what you do is you take care to place the responsibility where it actually belongs: with the financial decision makers. You start by finding a solution that will comply with their brief. Something like Tea's system above would be suitable. You get it down on paper, you cost it out to within a few hundred dollars, and you write down as honestly as you can the strengths and weaknesses of the system. (I find that having a system description followed by short "for", "against", and "summary" sentences works well.)
Next, you do the same thing with a technically ideal system. For example, Athlon XP 1600, 256MB DDR, Intel NIC, and so on.
Then you do it again with some other combinations. Is it possible to consider, for example, good systems made affordable by re-using some existing parts - e.g., hard drives, CD-ROM drives, network cards, cases, floppy drives, RAM? Be creative with this: maybe there are fifteen systems that are not too bad still, and which, with the benefit of some extra RAM stolen from the old systems, or a second HDD to fit the swap file, or one of the better remaining CD drives, could be made presentable enough to do another three years. Maybe some of the tasks that these systems do are lower priority, so you could use existing systems for those, so as to preserve some budget dollars for more critical tasks.
And so on. You want to wind up with a two to three page document with no less than three and no more than six options on it that you present to your superiors on the following basis:
The provision of IT equipment is a key decision for the future of [this business]. The upcoming purchase will be something we here at [name of business] will have to live with for the next three years. Always, with a complex system such as a computer network, there are trade-offs to be made: no decision is compromise-free. While I have been tasked with the responsibility for judging the technical suitability of the purchase, the decision necessarily also involves longer-term business strategy decisions which are more appropriately made by [insert name of decision-making body here]. This paper sets out the main purchasing options, together with a summary of the advantages and disadvantages of each.
Then you lay out your four or five options. I'd have something like this:
Option 1: 60 integrated systems as outlined by Tea: Celeron 1100, on-board LAN, etc. $350 per unit.
For: Low initial cost, complies with networking requirements.
Against: Obsolescent technology, shorter expected useful working life, minimal residual value.
Summary: A good solution but only in the short term.
Option 2: 60 seperate component systems based around Athlon 1600 DDR, Intel NIC, stand-alone video card, good quality case. $450 per unit.
For: Very much more powerful (and hence longer-lasting) than Option #1, more flexible to cope with unexpected new requirements, cheaper to maintain, excellent upgrade prospects for lower long-tern cost of ownership.
Against: Initial cost. Given contractually-required replacement in three years, we may face having to replace machines that are still perfectly suitanble.
Summary: The best technical solution, and quite possibly the lowest long-term cost solution, but more expensive in the short-term.
Option 3: 45 seperate component systems as in Option #2, upgrade and refurbish the 15 systems in Room 14A (give details).
For: Blah blah blah.
Against: Blah blah blah.
Summary: Blah blah blah.
And so on.
Business are like the military. They have a chain of command. And one of the most important things about the chain of command in an efficient, effective business, just as it is in the military, is that there is a productive flow of information in
both directions.
You need to know what the strategy of the business is (as it applies to IT purcasing] and
they need to know what the tactical constraints are that you, their representative, face in trying to implemernt their strategy successfully. They need to be kept informed of what the
real, practical effects of the IT purchase will be.
Sure, you, as a teacher, would love to have 120 XP 2800s with 2GB RAM and a 30-processor Opteron server. And they, as financial controllers, would like you to get by with a half-dozen Pentium 150s. It's not for you to say "we need such-and-such". But
is part of your job to make sure that you communicate the actual, practical import of your purchasing decision to the management, so that they know exactly what the consequences of the purchasing decision are. If, in full possession of the facts, they tell you to rush out and buy Option #7, which was 60 El-cheapo PC Chips crap boxes and that they don't mind if 30% of them are unservicable 90% of the time, then that's OK. You have given good advice and done all you can. It's not for you to
decide strategy. Your task is to
divine the strategy, and then provide the best practicable tactical implemetation of it.
