Some views of my journal do not allow back-dating of posts. This entry should be dated 2005-10-06.

Queensland Holiday, days 1—2 — sunburnt and stung
I'm two days into my holiday and dive trip. Our first dive was short, and included some time in getting the proper amount of weight for my new wetsuit. By the second dive we were feeling more confident, and by the second day things were just superb.

We've been collecting some great photographs, although we forgot to take the camera on one of the best dives so far, a massive 25m high bommie with crystal clear visibility.

So far I've discovered two important things, albeit a little too late. One is sunblock, I now have two very red shoulders where my wetsuit doesn't cover me. The other is the use of stinger suits. During today's night dive I got stung part-way through, giving me some rather-painful looking marks to show off to the rest of the passengrs and crew when I returned. As much as I hate wearing a stinger suit, the idea of being covered in nasty red sores is even less appealing.

Tomorrow we're diving the Yongala, an almost 100-year old wreck in an exposed position. It's often touted as "Australia's Best Dive", and all the people I've spoken who have dived it before seem to agree with that statement. Apparently the wreck is open to rocky seas and strong currents, and with our planned dive taking us down to at most 28 metres and then working upwards, I expect my air consumption will be pretty high. Unfortunately we only have 12-litre tanks on board, filled to only 200 bar. I've been spoilt with bigger tanks and 250 bar fills.

Food on the boat has been fantastic, and I think us vegetarians are eating better than the rest of the passengers and crew. Jacinta and I have a lovely double-cabin. The only downsides, besides from the small tanks, have been the bathrooms: there are only two of them, and they contain both a toilet and shower. One can spend quite some time waiting for someone to finish a long shower, even if one's personal needs are quite short.

Our equipment has been holding up very nicely, although there have been a couple of hiccups. I accidently lost my surface marker buoy on one dive, but another diver thankfully recovered and returned it. Jacinta also had a personal marker light flood on a night dive. It's probably our cheapest and least important piece of equipment, but salt-water and batteries sure make an awful mess! We still may be able to repair the light, but it will probably require quite a bit of scrubbing to do so.

Our Yongala dive is at 6am tomorrow, and I've been told one of the highly coveted showers is free, so I'll write more later when I can.

Some views of my journal do not allow back-dating of posts. This entry should be dated 2005-10-04.

Queensland Holiday, day 0 — travel
Today is my travelling day on holidays with Jacinta up to Townsville, Magnetic Island, the great barrier reef. Holidays are always an interesting affair for me. As I seem to be chronically short of time (or endlessly able to use it), I'm always concerned that holidays will take me away from more pressing tasks, the most significant of which is the completion of our Web Development course for later this year.

I now more fully appreciate why there is a 'holiday season' in December. If everyone else is taking a break, then businesses can more safely close their doors for a few days, or manage without key staff on-hand. Going on holidays during October always has me thinking about how we will effectively manage any important business events, such as course bookings, while we are away.

Despite all the worry beforehand, I'm certainly looking forward to the next few days. Diving on the southern end of the Great Barrier Reef was fantastic, and I have every reason to believe that the northern end will be just as good, or better. What I don't know about is the boat. I've been on some absolutely fantastic boats with great facilities, people, and food (Big Cat Reality immediately springs to mind), but I've also been on other boats that were less good.

The other possible hitch is our wetsuits. I have a 7mm two-piece suit, designed to keep me warm and comfortable in Melbourne, with chilly lovely 10-20'C waters. Our trip will have much warmer surface temperatures, and 20-30'C waters. I'm fortunate in having the two-piece, I can choose to wear only half the suit if it's too warm. Jacinta has a heavy single-piece suit with very serious seals. It won't surprise me if Jacinta decides to go diving in her lycra-suit instead if the water is warm enough.

Currently we're delayed in Brisbane airport, with morning fog delaying our arrival, as well as our connecting flight. Luckily I've borrowed a good book from Ian on the history of the calendar, which besides from providing entertaining reading is also answering me a number of nagging questions I've had for some time, such as how did we end up the days of the week (they're named after celestial bodies), why do they have strange names in English (some were renamed after Norse gods; Thursday is Thor's Day), and how did they end up in the current order (a rather fascinating system of naming hours after celestial bodies, with days taking their names after the first hour). The book also notes that 2012 is the end of the current Maya Great Cycle, when it is said that all things will cease to exist, and the world will be made anew.

Batteries and Dive Computers
I'm going on a diving trip soon, so last weekend was spent testing and packing gear. While testing the dive computers we discovered that my computer had lost some battery charge; not enough to cause any immediate problems, but I don't really want to find myself with a 'low battery' warning out at sea. It's not surprising that the battery has lost some charge, I've done a lot of diving with this computer, and the battery is holding up very well compared to the expected lifespans listed in the manual. Since the idea of changing batteries part-way through the trip was not appealing, part of my pre-dive work was to change the battery on land.

The recommended way to change the battery on this model of dive computer is to purchase a 'battery kit'. This includes a new CR2032 lithium battery (valued at $4.00), a new O-ring (valued at $0.05) and a new plastic battery cover (valued between $15.95 and $45.95, depending where you buy it from). The battery kit is a good idea, it ensures that you're not reusing worn, damaged, or dirty parts.

Unfortunately, finding a battery kit proved to be difficult, and I wanted my battery changed before I went away. Given this is the very first battery change, and that I have silicon grease and lint-free cloths at hand, I decided to just purchase the replacement battery, which are cheap and easy to find. What was interesting is some of the discussions I had with equipment technicians regarding dive computer design.

It appears that most dive computers have their actual electronics embedded in what's informally known as "potting mix". This is a waterproof gel which means that even if water does enter the computer case then it's unlikely to cause immediate failure. In fact, the most likely problem to be caused by water leakage is corrosion of the battery contacts. While that's a bad thing, it's a lot less dangerous or expensive than total computer failure. There's even a possibility that the computer can partially self-diagnose; the corrosion reduces voltage available from the battery and results in a low-battery warning.

The net result is that I need feel much better and safer with my newly replaced computer battery, instead of being concerned about damaging seals during the replacement and having to use a bottom-timer and tables during my holidays.

$chickens--, a case study in physical security
I had thought that our fully enclosed chicken run would mean we would be free from successful fox attacks. This morning, at around 5am, I discovered that was not to be the case.

It strikes me that our situation is almost classical in the security world, except with chickens and foxes rather than machines and h4x0rs. The attacker only needs to find one hole to get through, so the defender needs to find and close all of them to be secure. When an attack is successful, it's important the vector is found and stopped, otherwise it will continue. Sometimes discovering the vector is hard.

As best as we can tell, the fox (after months of unsuccessful attacks) now has an excellent grasp of the physical properties of chicken wire. It discovered one location where the nails holding the wire to the fence were further apart, pushed its way down vertically, deforming the chicken wire. Once inside, it grabbed a chicken, and then appears to have squeezed out by pushing and deforming the chicken wire on flap designed to allow leaves to be washed off the coop's roof. No wonder we have the saying "as clever as a fox", and that foxes have caused such massive destruction to our local wildlife.

Today we're patching up the holes, adding more tensioning wire, and adding more nails and cable ties so there will be no locations that a determined fox can squeeze through. Will that fix the problem? Yes; for a while, until the fox discovers a new and completely unthought-of means of entry.

The correct solution to this would be "defense in depth", but even there we see the classical problem of security vs convenience. We could look the coop every night (as well as the run), but then we have to visit the chickens every day at sunrise to let them out. We could buy a dog that sleeps in the yard, but then we'd have to look after the dog. We could install motion-activated floodlights, but then we'd have spend quite a bit of time and money installing them, and the fox could just get used to the light. We could develop an amazing fox-killing robot warrior, but there's probably a council ruling against it. We could deploy a large moat of steaming acid, but then we'd need to refill the moat after rain and hot weather.

We could always buy our eggs from someone else, but I would hate to admit defeat.

O'Reilly Book Distributors
One of the problems with running a training business is getting one's customers to pay their invoices. It's not like we've had any bad debts, it's just that most of our clients are from "big business", and for an invoice to be paid it needs to be posted, sat on someone's desk for a month, mis-filed, resent in triplicate, routed to the wrong accounts department, resent again with a supplier contract, forwarded to the legal department because there may be IP issues, who forward it to risk management because there may be public liability issues, who forward it to HR because it involves training, who forward it back to accounts for payment. Eventually, some poor manager asks to pay it on their personal credit card so they can have it reimbursed later.

We tried to encourage prompt payment by offering a discount, but this ended up being more work as we had to issue a second invoice when the first wasn't paid in time. We'd then go through the same proceedure a second time.

Eventually we discovered a trick. Offer free books if the invoice is paid on time. Because this directly benefits the employee, the invoices start to become hot stuff, and we're seeing a lot more invoices being paid on time. That's a great thing, as it means we can spend less time chasing accounts, and more time writing new courses and presentations.

Unfortunately, there's just one small speed-hump, and as our business has grown the the speed-hump has got bigger and bigger. It's the actual process of ordering the books themselves.

We have an agreement with the Australian distributor of O'Reilly and many other technical publishers. We order the books with a significant discount, and they arrive either at our office or one of our training labs. Everyone's happy, or at least that's the plan.

In reality, our distributors have become increasingly more difficult to work with. We get a lot of conflicting information about whick books are 'in stock', or when books are due to arrive. We'll be told that all of our books will arrive at a training centre, only to discover that some are missing and have a 4-8 week delay.

One frustrating situation included the books being returned due to courier error. Rather than us being notified of the problem, the books were re-sold to another customer. It was only when we called to ask why our delivery was so overdue that we were informed as to what happened, and that for us to get the books we would expect another 4-8 weeks delay. We still don't have those books.

I'm at the point where I have no confidence that we can supply any of our customers with books in a timely fashion, particuarly if it's something that has been recently published such as Perl Best Practices. I don't mind if books come late, as long as we're told before-hand, rather than discovering on the day of a training course. Our distributors are quite aware that we're unhappy, but have not yet demonstrated any plan to rectify the situation. Apparently giving us a phone-call to let us know there are delays is completely out of the question.

The pain of ordering books is now starting to approach the pain of chasing accounts. Unless we find a new distributor (or our existing one lifts their game), we may have to reduce our book freebies to the two books we can obtain reliably (Programming Perl, and Perl Coobook), or worse still give away boring things like T-shirts or game consoles. I'd be very sad if it came to that, as I'd rather our freebies generate revenue for some of Perl's best minds, rather than for textile or electronics manufacturers.