I don't know how many of you have had the pleasure of an infestation of the above, but let me tell you that I've never been so frustrated and ready to give up and re-install Windows from scratch as I was late last week.
I've never really had a problem with spam or viruses before - I'm careful with what I open, keep a low profile on e-mail and so on. But a particularly malicious variant of the Sasser worm (one that finds you, then auto-executes) got hold of my laptop last Tuesday overnight.
It took control of parts of my system and my browser. It deleted the Google toolbar and installed something called the Elite Toolbar instead. It hijacked all my searches and put up ad-related links instead. It kept polling a site I think is in China to download "payloads" for the Trojan-horses it was running. Every time the PC rebooted, it would shape-shift into something new, hiding somewhere else (system folder, temporary internet files, my documents - you name it). I had pop-ups galore, many of which I found were actually living on my own PC and running on the clock.
I spent a couple of days taking the measure of the beast. I Googled around on what traces it left, killing what I could see to be rogue software. In the process, I found some pretty useful software available for free on the web, and I got some sterling personal assistance from one of the gurus on Spyware Info.
The weakness in my PC was due to having recently connected this laptop directly to the broadband connection. Previously, it was on my home network, behind a firewall. But my laptop itself was running neither anti-virus nor a firewall when I hooked it up. It didn't take too long for this sleaze to take over my machine.
I bought a copy of Norton 2004 for anti-virus. It's okay, but it just can't see a lot of the spyware stuff, because the spyware stuff is designed to hide from it. And it's crap at solving spyware problems once they're manifest. It takes forever to scan the PC and for things it can't delete it makes you go into Safe Mode. I've spent a lot of time in Safe Mode this past week.
So, I used Ad-Aware in the freeware version available on-line. It found more of the spyware stuff, and told me where it was hiding, but it was not able to rid me of the problem definitively since it kept coming back with every re-boot.
For that, I started running another freeware program called Hijack This. This scans your PC and tells you what's running. By comparing it to what should be running, you (or an expert) can tell you what's malicious and point out how to kill it.
That's where the guy on the "malware" forum at Spyware Info helped out. I posted my HJT log and he held my hand through the process of shutting down the rogueware. I'm not sure it's all gone yet, but it's certainly been neutralised.
And after all that, including lessons in things like editing the registries, I've done what I should have done in the first place and downloaded the latest version of ZoneAlarm, locked down my browser security and am in the process of installing Windows XP Service Pack 2.
I don't purport to being a malware expert, just wiser for the experience. I'm posting this a sort of public service in gratitude for the excellent support I've received (for free) and in the hope that Google might pick it up and point others to the resources I've discovered.
A timely article in today's Guardian, Cost-cutting 'jeopardises North Sea rigs', points out some real-world examples of the impact that cutting back on experienced manpower can have on reliability, safety and the environment. I was on about these from a similar perspective last week.
Fears are growing that safety is being compromised in Britain's North Sea oil and gas industry, and there are claims that corners are being cut to reduce costs.
À propos the trend of cutting costs and pushing margins to keep the pots and pans just ticking over, one man is quoted as saying:
They are trying to cuts costs on old installations (which are) falling into disrepair; plant and equipment is failing on a regular basis, systems are becoming old, valves are passing. They are pushing maintenance right to the very limit.
and here is an industry specialist talking about cutting head count, which, the article reports, has dropped by 12,000 workers over the past decade:
"It is no different from any other industry where the move over the last few years is always to get rid of people and try and operate on fewer," said Colin MacFarlane, professor of subsea engineering at Glasgow and Strathclyde Universities, and himself an offshore contractor.
"The offshore staff are already stripped down. There is a very strong feeling that there are unseen hazards. You cannot quantify it, but we believe they are going up because the numbers employed in maintenance, the numbers involved in operations have been dramatically reduced and when you are talking about maintenance, that is scary. There is a stage where maintenance falls where nothing can be done to make things better. It may be that companies believe their increased focus on safety management is countering this, but recent incidents suggest not."
It's that last bit that scares me. If you want to push the operating envelope beyond its design, you probably need more, not less, human expertise in the loop. More judgement by experienced workers. If you cut your numbers to cut your costs while at the same time pushing that envelope by extending your maintenance intervals, something will eventually bite you, quite possibly catastrophically.
Why? MacFarlane nails it: "You cannot quantify it." Because the methodology wants to push the limits based on prior experience, but the old mutual fund maxim "past performance is no guarantee of future results" has to apply. We don't have enough good data to confidently push beyond a certain limit. And by cutting headcount, we've just cut back on our experience base.
Where there is good data, if you are respecting good engineering judgement, you should be okay, but if you're doing this while at the same time encouraging a culture of under-reporting and corner-cutting, you are undermining the remaining safety checks built into the safety management system.
Sad to say, but maybe the ones who are "not required back" are the lucky ones.
A colleague of mine gave me a copy of this article to read. It's from a trade journal dedicated to a fairly narrow specialisation in my industry - the Process Engineer, and particularly the Control Engineer. These people are customarily responsible for the day-to-day operation and optimisation of industrial plant equipment. The article is long and suffers from some bad editing, but it's worth a read.
What I found interesting about the article is that here, for once, the author is not simply blaming off-shoring or outsourcing as the culprit for the loss of jobs in the engineering profession. These trends are visible, but they are only part of the big picture. This meshes with my own experience. Since these engineers are so specialised and must normally be present at the plant location, it's not that easy to farm out their work. Rather, industry is just deciding to do away with their role in the organisation.
The author, Rich Merritt, looks at some of the bigger trends in industry. It left me looking at my own career progression and thinking about some of the traditional assumptions we engineers tend to make about our importance in the process environment.
Chief among the trends is the growing importance of non-engineers in the optimisation of installed capital:
Part of the problem appears to stem from the fact that engineers and chemists no longer play as important a role in managing and operating process plants. The accountants and lawyers hold sway now, and have less of a professional stake in the work process engineers do (emphasis added).
I watched this trend pass through my industry in the mid-1990s. Industrial companies used to be largely run by the men who built them. You could look at the Fortune 500 list and see lots of chemical engineers among the industrial firms. These were the men who knew every nook and cranny of the operation. The senior management were lifers who would follow a progression from process engineer to area supervisor to operations management to general management. And they could keep the place running with duct tape and spit, as it were.
But there was something wrong with this picture, and that was the one-size-fits-all approach to management. The assumption that if a process engineer can run a plant, he can run a company. This tends to build companies full of process engineers! That's been the situation I've encountered in nearly every conventional organisation I've worked with over the past 15 years.
But some time in the 1990s, the mantra of shareholder value started to gain the upper hand. Maybe guys like Welch and Bossidy had something to do with it. Maybe MBA culture. It was obvious that senior engineers were good at what they did, but look what they cost! Wasn't it inevitable that as other costs were squeezed, the lens would turn inward on the cost of the engineers themselves? Suddenly engineers were no longer essential. The game of return-on-capital-employed (ROCE) is about reducing costs for the same asset deployment. If you're not planning to employ less capital, you can only improve the return on what you've got by cutting the costs associated with it.
Companies started to think in terms of ROCE. Some big firms decided to sell off marginal assets to improve their balance sheets. Unocal sold off its refineries altogether and decided to focus on higher-value offshore operations. BP slimmed down its refinery base and farmed out its in-house engineering functions. The capital employed by the big guys was written down. These assets were ideal for the new financial players (Tosco, Valero): pay pennies on the dollar of installed cost and run a very lean ship. The new breed had low capital employed and in order to run marginal assets profitably, they had to run with extremely low costs. Engineering department overheads were slashed. Then the big guys began to see that this seemed to be working.
At the same time, there has been a trend toward more high technology in the process environment. Basic technologies melded (e.g., computers with controllers), and synergies emerged (APC, real time optimisation, expert systems) that allowed more automation and optimisation without requiring the steady hand of the plant engineer to keep things ticking over. If you needed engineering services, you could buy it in (so, indeed, some of this operation was outsourced). Your central in-house engineering function became that of deciding what services were needed and buying them at the lowest possible price.
The article points to a bit of the vicious circle phenomenon, where overhead was squeezed out of an organisation, taxing the remaining resources to the point where they were no longer able to look critically at their own function and make improvements, which in turn made their role somewhat redundant and less valuable to the organisation overall.
There's a book on this subject that I'd like to have a look at some day: Slack: Getting Past Burnout, Busywork, and the Myth of Total Efficiency, by Tom Demarco. I haven't read it, but the key point is raised in this Slashdot review. Companies slip from peak efficiency when they squeeze the human factor too much.
The question comes down to the lens through which you view the organisation. From within, the engineer looks at the millions of dollars of equipment for which he is responsible. He looks at his salary, looks at the cost of suboptimal operation and says, "Darn straight I'm worth my keep." I've certainly looked at the world with that view before. When he looks at an org chart full of engineers, he feels pretty confident in his own value to the organisation.
But the MBA or accountant looks out and says, "Just how little can I spend and keep my pots and pans ticking over?" You don't pay bankers on the basis of the money they handle. You pay the minimum number you need to keep your business running. Non-essential functions slip aside. This "pay-later" strategy seem attractive in an environment of minimising cash outlay. And I will say that it can work - I've seen it do so.
When the organisation loses is when optimisation is considered a non-essential function. If you don't know how much it's costing you, you can ignore it. These days, you can buy benchmarking services that tell you how you are doing in relation to your competitors, so you don't need your own people to tell you. And that's valuable, because the worldview of your own people is primarily influenced by what they see within the fenceline. And you can also buy services of companies like mine to help you get closer to the optimum. Hiring in services is attractive, because you can look at them in terms of their value propositions. But, as the article points out, you are not likely to get the same level of experience that you would if these were your own people. When is that a losing proposition?
In my experience, the balance was tipped when short-termism came into play - the idea that period-on-period (quarter-on-quarter, for instance) improvements in efficiency (particularly in efficiency of ROCE) were necessary to drive shareholder value. This ignores the impact of longer-term projects that take time to design, implement and operate.
Engineering is all about usefully harnessing pure science. Engineers always had one eye on the financials, but they also had one eye on the plants. When you separate the technology from the finance, you are maybe gutting the core identity of the engineer.
Am I dewy-eyed here? Not really. Like I said, these days I make my living by providing optimisation services as a consultant to third parties. One of those career conundrums. The engineer in me says, "Look, it's obvious, the process engineer is worth his keep." And yet the consultant in me says, "Hey, this is how I make my living. Keep it up you greedy SOBs!"
I was playing around with the streaming "radio" stations in iTunes this weekend and came across a stream from something called "Magnatune". I followed a link to their homepage, and I have to say I was very impressed with what I found.
Here is a company applying the "shareware" model to music. They offer up their content free (at mp3 quality) for non-commercial use. If you want a better quality version, or if you want to support the artists, you are free to contribute (the range is $5-18 per album, with $8 suggested). And from any money you contribute, half goes to the artists.
I thought this was a brilliant idea. And then I found they actually have music that I've been meaning to buy. They have two albums by The Dufay Collective, a group dedicated to performing music from the Middle Ages and Renaissance. I had heard some of their recording of "Songs in Praise of the Virgin Mary", 13th Century Spanish cantigas, on the BBC some years back. This particulary recording is not available from Magnatune, but I did download their Cancionero recording. First in mp3s, then, after I bought the CD (on-line), I downloaded as WAV (CD quality) files (it takes me about 5 hours per CD with my limited bandwidth - just perfect for an overnight download).
I also bought a recording by a group called Stellamara, and I've been experimenting with some of their other artists - check out the recording by the Lavra monks in particular. If you go to the site, click through "license" and then select the "non-commercial" - you just have to confirm that your use will be non-commercial and you are free to download any / all of their content.
Such potential - you can sample everything and buy what you like / keep. And the artists publish their music on a Creative Commons license, are not bound by exclusivity clauses, and get paid at probably 400% of what they would get from a major label. I'd be happy to buy more music on terms like this.