Next moves

I was thinking about the future today. To make any reasonable judgement about the future, one must demonstrate some wisdom with regard to the past.

When I was in my twenties and dissatisfied with my first few jobs, I can remember thinking that I was simply experiencing a bad run of luck. All my jobs had sucked in some way or other and if I could just jettison this lousy company, all would be well, because my next job would surely see me at a company where people do things the right way and I wouldn’t have to be frustrated by the futility and wastage any more.

Sooner or later, though, you realise that it’s impossible to roll the die a hundred times and see the 1 come up on top every time. The law of averages has to kick in somewhere, right? Rather reluctantly, one must ultimately conclude that inefficiency, wastage and futility are not the diseases of an ailing company; rather, they are endemic to the working environment itself and all companies suffer from these ailments to some extent.

A problem with the field of system administration is that it’s hard to convince people of the necessity of good procedural methodology. All too often, young managers mistake their strong intuition for actual experiential wisdom. In other words, when they tell you which approach to take to a problem, be it technical, political or whatever, frequently they think they are drawing on experience in a similar situation, when really they are simply following a hunch. Balancing self-confidence with a healthy sense of one’s limitations is quite a feat of acrobatics for many.

What I’m getting around to saying here is that there will always be particular kinds of frustrations when one is in the employ of another. One can either resign oneself to these frustrations, seek another line of work, or seek to liberate oneself from the yoke of toil in the service of another.

All of which simply brings me back to the beginning of this entry: I was thinking about the future. With the prospect of some modest financial reserves at my disposal within the next year or so, the notion of becoming my own boss becomes a very real and attractive possibility.

But what to do? If you are going to start a successful company, you need a good idea, funding, a certain measure of luck, conducive market conditions and expertise. The most important market condition is that there is a genuine need for the service or product you intend to offer.

I’ve always wanted to write a book related to the field of system administration, but the biggest problem there is identifying a gap in the market and coupling that with the need for expertise. It’s hard enough to come up with an area of the field that hasn’t already been published to death. If you succeed, you need to be pretty lucky to also happen to have strong expertise in that area.

I could write a bad book on several different topics. I am knowledgeable on all of them, but my book would not stand out from the crowd, because I am not a true expert in those fields. There are one or two areas in which I am fairly strongly skilled, but those areas are already represented by strong books covering the field. And that’s why I haven’t yet published a book: because I don’t want to be a mediocre author.

Anyway, that was another digression, simply because it’s analagous to deciding whether it makes sense to set up one’s own company.

I can think of a massive gap in the market: a decent calendaring system that runs on Linux. Decent is a subjective notion, of course, and I won’t go into it here. Suffice it to say that all solutions I’ve seen in the last ten years have been inadequate in one or more ways, sometimes hopelessly so.

So, why not set up a company to produce a decent calendaring package for Linux? Because I don’t know enough about it. I’m not a talented software engineer, so I’d need to find someone else to do the actual work. I think a company has its greatest chance of success when started by a founder who, himself, has actual expertise in the field into which he is entering.

How about Linux consulting? Well… A packet-filtering firewall for customer X; a Web server for customer Y; e-mail infrastructure for customer Z. It’s all very nice, but it’s yesteryear for me. Been there, done that, etc. It just doesn’t excite me any more.

Perhaps that lack of excitement is burn-out; perhaps I’m simply jaded with my profession. I suppose I won’t really know until I’ve been out of work for several months and able to recuperate. I’ve met some great people in my career, however, and the thought of starting a company with them is very appealing. Of course, they, themselves, would have to be amenable to the idea, too, but I already know of a couple who would be.

So, it’s really just a question of coming up with a good idea and being knowledgeable enough about the associated field to have the confidence and wherewithall to succeed. Easier said than done, though.

While on the subject of calendaring software, I released version 0.2.1 of Ruby/CorporateTime tonight.

This entry was posted in System Administration. Bookmark the permalink.

3 Responses to Next moves

  1. Geoff says:

    If youre not knowledgable enough about a calendaring tool to write one, why not research it and GET knowledgable about it?

    And while youre doing that, write a few throwaway ones to help figure out how things work for the internals.

    No one learns how to do this stuff by osmosis. 🙂


  2. Ian says:

    Principally because I’m a system administrator, not a skilled software engineer. And calendaring is a pretty tough nut to crack. If it weren’t, most calendaring solutions out there today wouldn’t suck as badly as they do.

    I’m not saying it’s impossible, but I’d need to feel a huge affinity with an idea to invest the kind of time and energy you’re talking about. I’d more likely be better off picking something I’m already good at and attempting to be the very best in that field.

  3. Geoff says:

    Yeah, cause NO one ever does a crappy job at things when theyre making it.

    Option 1) Calendaring is hard. Like going to the Moon.

    Option 2) Lots of software is crappy, very few software is good. Because of corporate culture, disatisfied employees, lack of interest in their product and disorganization in a company leading to even good things turning bad (like selling the company to another and it being ignored for years).

    If you want more time with Ruby, you have to pick projects where youll be using it a lot right? 🙂


Leave a Reply

Your e-mail address will not be published. Required fields are marked *