Tony Weir


Biography has not been added


's contributions
    • Yes, I really don't know what point the author is trying to make. The main argument against assembler is that it's non-portable. Processing power and memory is so ridiculously cheap today that there is no longer any argument for using assembly for performance reasons (although even five years ago I wouldn't have said that's ALWAYS true). Of course a good programmer should have some understanding of the CPU's assembly language and architecture - if only to hand-craft things like startup routines and interrupt handlers. Beyond that, it's pointless.

    • I think this can be summarized as: too many people pretending to be engineers; not enough REAL engineers ;) As for this comment: "All of these new computing models require architectures that are very different from those that went before, and what older folk learnt in their engineering schools and training programmes." You have to realise that the attitude to education, especially in Asia, is very different to The West. University is where you go to sit at the feet of great sages, and shut up and listen while they funnel (outdated) knowledge into your head. You then go to a blue cubicle and never learn anything ever again. It's the main reason American and European engineers still have a significant edge over those in (say) China.

    • There is no such thing as 'self-documenting code'. Programmers who can't be bothered to put in a few lines of explanation are not 'real programmers' - they're just lazy and incompetent.

    • The Old Man got it right, but even the most elegant solutions need - as you say - a bit of an overview to at least explain what goes in, what comes out, and broadly what happens in the middle. Far too many programmers can't be bothered to write even that much - maybe no more than a couple of paragraphs for a substantial function.

    • Excellent article. It drives me crazy when I have to deal with code which is full of badly-spelled, ungrammatical comments or documentation which makes no sense at all. The authoring company being foreign (ie., writing the docs in a foreign language) is no excuse whatsoever: hire a native speaker. Worse, of course, is software with NO COMMENTS AT ALL. I recently acquired a copy of Inferno - the OS originated at Bell Labs - and was shocked to find the code not only of dubious quality but also essentially uncommented. I don't have the time or inclination to figure out what J Random Hacker was thinking when he wrote that stuff - which is a huge pity, because Inferno has enormous potential.

    • Interesting about the divorce rate. Any theories about that? Can't be the engineer's innate desire to solve problems - if there's one thing women hate, it's a man who always has solutions to their problems. Maybe it's the fact that they stay safely locked away in their basement labs - the wife can just pretend she's single.

    • Engineers don't get respect because they don't expect it. They too often defer to managers, politicians, and other "experts", who have drilled it into engineers' heads that they are social inadequates who can do least damage if kept inside their cubicles. Movements such as Engineers Without Borders are a direct reaction against this kind of thing. I am involved with an engineer-led organisation designing carbon-negative infrastructure component (yes, carbon-negative) for sale in developing countries. Why? Because the politicians, the managers, and the pundits have so little engineering knowledge that they don't understand that such a thing is possible. Therefore they won't do it. The politicians have their place - they have the people skills (engineers are social inadequates, remember?) and the PR savvy to get the interest and the funding flowing, but they need to understand that it's a team effort; without the engineers making things happen, they're just a nobodies with a nice suit. If engineers want respect, they need to learn to strut.

    • I think a recession is sometimes an excellent time for a startup. Americans are just having a hard time affording regular donuts and SUV fillups. To paraphrase Paul Hogan, "that's not a recession. THIS is a recession" : 70% of the planet's other inhabitants are starting wars over water rights and worrying about how they will eat or keep the lights on. With foreign investment hard to come by, anyone with a useful contribution to make to solving these problems is onto a winner. Ray Keefe is correct - if you're part a team with the drive and passion to stay chained to the bench, it's possible to self-fund and present VCs with a full-fledged product (in my experience, they're more impressed by something that works than a powerpointless presentation). This is exactly what we're doing, and it's working out just fine.

    • Of course lying to your boss is a stupid thing to do - but fudging schedules isn't lying. We do it for two primary reasons : firstly because we often don't KNOW how long a project will take, even to a first approximation; secondly, we're rarely given the time to actually figure out an accurate schedule; and thirdly, things like feature creep, ignorant bosses or sales teams, and trade shows are all a valid part of the schedule calculation. Work doesn't just involve sitting at a desk and coding; we have to answer phone calls, speak to bosses and other team members, and generally deal with everyday wibble that business involves. A boss who routinely cuts schedules by a known fraction is simply part of the equation.