Posts by BenWilson

Last ←Newer Page 1 2 3 4 5 Older→ First

  • Envirologue: Choose Wisely, Grasshopper:…, in reply to Bart Janssen,

    You are being obtuse for the sake of argument. That is not helpful or productive, it is simply a waste of everyones time.

    You always say that, and it's always wrong. I'm almost never obtuse, except for humorous purposes, which are clearly signalled, for the very reason that it does waste time, including my own. It's quite arrogant to suggest that you know my motivations like this. I'm arguing in good faith about something I believe, the way I usually do. Please do the same and quit making it personal.

    The fact is that every fact you bring up has been the case for thousands of years, just with different names, faces and places. If National changed this, then at least have the intellectual honesty to see that they actually did something good as a result of an opinion poll. Perhaps the public called strongly for this based on evidence. Good on the public, and good on National for listening to the public.

    Your idea of poll driven policy is frankly stupid and no amount of waving your philosophy degree around is going to make it less stupid.

    I'm hardly waving my degree around here, in fact I haven't even brought it up, because it's not relevant. You ought to think about whether such a comment fits the spirit of this site. Poll driven policy is hardly a wild, outrageous idea. It's a fairly standard extension of the idea of democracy, getting around one of the main shortcomings, which is that people don't have time to answer every damned question. But a good sample of people can have the time, and you can get the benefits of democracy without so much of the shocking inefficiency.

    Furthermore, poll driven policy is not even limited to this government. It would be a fair cop on Labour in it's last term that there was a lot of it. It would be a fair cop on a great many democratically elected governments that they went to some efforts to actually find out what the population want, and it's a credit to them in some cases when they do. I highly doubt a lot of the progressive changes we've seen would have ever come if conservative politicians relied only on their own intuition and peer pressure to decide if something should be done.

    Auckland • Since Nov 2006 • 9036 posts Report Reply

  • Envirologue: Choose Wisely, Grasshopper:…, in reply to andin,

    And when the majority/influential minority of population is wrong, things can go very wrong.

    Sure, there's no form of government that is foolproof. What democracy has over the others is the extent to which the values of the populace in general come to bear on the choices made. I don't think there's anything inherently wrong with extensively consulting that value system. Quite the opposite, I think there's something very wrong with believing that the value systems of the populace should play second fiddle to some higher truth, as administered by "leaders" and "experts in politics". It's not like that hasn't been tried as a system - in fact most of human history was like that.

    But of course entire populations can be wrong. We probably are wrong now about a lot of things. But I suggest that on matters that are scarcely factual at all, like our value systems, it's quite hard for populations to be worse than individuals whose own hubris and wealth can easily bring to bear all sorts of considerations that care nothing for the way the bulk of humanity feels about things.

    It's not like the population is too stupid to consult experts when they're out of their depth. Ask a random person what's wrong with a broken car and they know that a mechanic has a better chance of working it out. What they might not accept (and rightly so) is that some generalist intermediary is better at getting the correct information out of the mechanic, and so they should always let the intermediary decide how their car gets fixed. And yet, that is exactly what our relationship with politicians is like.

    Auckland • Since Nov 2006 • 9036 posts Report Reply

  • Envirologue: Choose Wisely, Grasshopper:…, in reply to Brent Jackson,

    Leaders are supposed to lead. Not be buffeted about by public opinion

    You aren't going to convince me of that point just by claiming it again, or forming it as a truism. If an opinion poll showing that people now want Warrants of Fitness for rentals caused the government to decide that is a good idea, then I'm glad I don't have to wait until the next election for them to "lead" us into it by putting it in their manifesto and campaigning on it, before doing anything. Responding to public desire (which can itself also lag behind genuine need) isn't in itself a bad thing. Sure, deaths from poor housing could have been prevented. By National. And Labour before them, and National before them, and so on into the ancient world. But they weren't. At some point a public consciousness of an urgent need reaches a sufficiently critical point that change is a good idea, and I'm glad that being pigheaded about the purity of our electoral signalling as the way for the government to get it's mandate isn't stopping some good being done (so long as it actually does get done - I'll wait and see on this actual point).

    Only some people get polled. Only some people respond to polls. People only respond to the question being asked in the poll. And most importantly sometimes people ask for one thing but need something different.

    This is the more serious criticism of my point. But just take the word "Poll" out and replace it with the electoral process and you'll see that your points here apply in spades to the alternative.

    Only some people get to vote. Only some people do vote. People can only vote for what is on the ballot. And most importantly sometimes people ask for some kind of government, but need something different.

    We have SFA ability to signal through our votes, particularly on the actual things that do matter to us. You don't get to vote on whether we should have Warrant of Fitnesses for rentals, but you can answer an opinion poll on it. After the poll a level of confidence in how much people feel about WOFs for rentals has actual data behind it, whereas after an election it's still all speculation and reckons. We end up having to take polls just to find out why people voted the way they did.

    It’s relatively easy to do things that are popular.

    Not always. There are many popular propositions that have never been acted on, because there is more than just popular opinion influencing them. Sometimes there are strong institutional barriers, or influential minority interests, and sometimes the popular things are actually quite hard to do in and of themselves. Otherwise the public health system in the USA would long since have been fixed.

    It is a much harder thing to do the things that are right and even harder to do the things that will be right in the future.

    "Right" is way, way, way more arbitrary than "popular". You could personally just say that they've never done anything right, and we'd be arguing all day about whether that was true or not. But you couldn't easily say that something was popular or not without some decent evidence, and any evidence against would probably do a lot to sway you. "Right" is a value judgment.

    So comparing "popular" with "right" is like comparing apples to some unspecified other thing that isn't apples. It's not even apples to oranges, because we know what oranges are. But what "right" is is simply ... undefined.

    Which is why it's not much of a guide to say that the government is there to do the "right" thing. Of course they are. But what is that? In my opinion, it's not always even vaguely opposed to what is popular, since that in itself is the popular judgment of what is right. It's only wrong to the extent that the population is wrong.

    Auckland • Since Nov 2006 • 9036 posts Report Reply

  • Envirologue: Choose Wisely, Grasshopper:…, in reply to steven crawford,

    This is where I’d like to be more quick at typing, and thinking.

    FWIW, I think you do fine.

    The obvios way to get the maximize market share, is to bring to avoid saying anything controversial, or to far outside of the square.

    Yes. Also, I find it hard to get hip to any system that has internalized a market economy so much that political parties would see themselves as companies and voters as their punters. It's not how I see my own political consciousness, as a consumer of big promises and lies, that I pay for with my vote. But, crappy as such a reality is, I don't object to such a "company" doing market research any more than I do for actual real companies. It's how they make their products better for their customers, at least to some extent. The side effect of them using it to get away with crappy product just by knowing how to target their punters better is a stink reality about the level of ignorance about the product that is widespread in punterland. Breaking through that veil of ignorance is more the work of third parties. Expecting the system to do it itself, for companies to have any interest in setting up an environment in which they have to work harder to get less sales, is not realistic.

    Auckland • Since Nov 2006 • 9036 posts Report Reply

  • Envirologue: Choose Wisely, Grasshopper:…, in reply to steven crawford,

    I too, would like to see changes to the electoral systom, that would lead to less poll influenced voting habits.

    Just to think outside the box for a second, I'd like to see changes that lead to more poll influenced politician behavior. I'd like more signal to pass through our triennial popular vote, but even with the perfect system I think we're already nearing the limit of what can be achieved by that. Even if every nuance of our little tick in a box is perfectly captured and meted out its proportional influence, it's still a system in which more than 70% of the people elected on the night will be from the Nationolaboural party, and they will spend the next 3 years picking and choosing amongst whatever popular (or unpopular indicators) that they like.

    I'm always a little bit baffled when people get bitter on a PM for following polls, like popular opinion isn't supposed to be what democratically elected leaders should be considering. The only part that embitters me is that they get to be so selective about which polls they do follow. But this process of becoming informed about what people want, using the power of statistical analysis is not, in itself, some terrible evil. It could just be less piecemeal and arbitrary, more formalized, more public, more open, less driven by rich and/or powerful minority interests. Then we'd be approaching a more participatory system.

    Auckland • Since Nov 2006 • 9036 posts Report Reply

  • Hard News: Campbell interviewed, in reply to Rich of Observationz,

    But teams aren’t randomly assembled. The company decides that it can import coders from the Philippines or someplace on a temporary work permit and pay them $20k a year – the good ones then realize they’re being screwed and jump ship to proper jobs.

    Yes, if you deliberately set out to select a team of shitty programmers, you'll definitely succeed. Alternatively you could take the opposite course of keeping the good ones, and pay them more.

    the bigger the team, the less efficient. If you can size a job to be done by a single developer in the required time, that’s optimum.

    Not in every case. You might want to get to market faster and a job that takes one person a year could take 2 people only 6 months, or 10 people only 4 months. The 2 months could be worth millions of dollars in sales, easily justifying 8x2 months extra wages. Also, I dispute that less are more efficient, even in general. Programming has many specialized roles, and you don't want the database guru sitting there writing code in a language he/she hasn't yet mastered just because you have a management theory, nor do you want the specialist in optimization algorithms wasting months working out how to set up the database. In general, synergy, specialization and economies of scale tend to make teams get stuff done a lot faster. And if your project is large you have no choice anyway.

    But certainly there are numerous inefficiencies in teams that a single operator doesn't have, and that lasts for as long as the operator stays there, and hasn't built up a large code base they have to support, at which point their idiosyncratic method that was not extensively reviewed or documented because it didn't need to be is a big liability.

    managers cost money. Less programmers (especially if they mostly self-manage) means less managers, hence more functionality per $$.

    Unless, of course, their lack of management screws the whole project, which is hardly an uncommon scenario. Someone is always managing them, even if it's the customers themselves. Which means they're wasting their own time on a task they could outsource to a professional, and taking a huge risk on something that's possibly well outside their competence.

    In production per dollar of wages, the expensive programmers cost less.

    That might be true in general.

    Of course, there are many reasons why management (and staff) might not want rock stars on high six figures in the company.

    The age-old solution is to promote the rock stars to some kind of management.

    Auckland • Since Nov 2006 • 9036 posts Report Reply

  • Hard News: Campbell interviewed, in reply to Rich of Observationz,

    Are you aware of the research that shows a 10x variation in productivity between programmers?

    No, but it tallies entirely with my 20 odd years of experience with programmers. Some of them have incredible output. In others it's hard to understand how they could even call themselves a programmer. As I said, most are in the middle of that range, although there is a long right tail to that data. The exact same comment could be made of many highly skilled professions. Of course there are outliers in both directions, so we can have an order of magnitude difference between the top and bottom. But you'll notice that the mean productivity in the link you gave is around the 4x mark, and the interquartile range about 2.5x to 6x. Half of the programmers fit in there. It's not going to be hard to find people who are 2.5 to 6 times better than the worst programmer on the books (the baseline of 1, although I'd say that is a false baseline since the least possible ability is actually zero).

    (A corollary to this is that since the best programmers are generally self-managing and productivity declines with team size, two really good programmers can do the job of many more than 20 average ones, and require a lot fewer managers

    No, that's completely false. two exceptionally good programmers get 20x. That's slightly less than 5 average ones. They're only the equivalent of a team of 20 nearly useless programmers, who are themselves also completely exceptional, the chances of randomly assembling such a team are astronomically low (on the order of 0.01^20 if the worst person is one in a hundred). It's not exceptional to be unable to program, most people can't. But it's exceptional to get and hold onto a programming job if that is your skill level - 4 times worse than average. Even management will notice.

    Good programmers are harder to find

    I don't think so. People who are trained in it are mostly about average. Half of them are better than that. Exceptionally good programmers are hard to find - by definition. If by "hard to find" you mean "rare". If you mean "hard to identify", then I don't think so. That's only the case for n00bs with no track record, who are, mostly not yet good programmers anyway, even if they might become so later. If you mean "hard to get to come to a job interview" then that's actually mostly a statement about supply and demand. Of course the guns want to get paid heaps and are probably locked in by some lucky employer elsewhere. What's hard isn't finding them, it's paying for them.

    I think most of the same comments apply to management. The difference is that management quality assessment is a much less precise thing than measuring programmer productivity. It's no wonder they end up using "the bottom line". Because all the other KPIs in the world can be considered total rubbish if the bottom line is killed by them and the company dies as a result.

    Auckland • Since Nov 2006 • 9036 posts Report Reply

  • Hard News: Campbell interviewed, in reply to Bart Janssen,

    Should it be rewarded more than what it is most time (doing chores)? No.

    If that is all a particular manager is doing then I agree. But "doing chores" doesn't necessarily capture what the role of management is. I'm presuming you mean the tedious keeping of records relating to team goals and doing minor duties that could as easily be performed by a secretary, then I'd agree, but in most of the places I've worked, the good managers usually do actually get the secretaries to do that, or the workers themselves, or by improvements in the systems and software. This frees them up for other more fitting duties like planning, promoting, communicating, analyzing performance, looking for improvements, hiring, firing, conflict resolution, getting customer input and feedback, selling the team's output, advocating for more or less or better work, keeping informed about the intended direction etc. All of these tasks could also be separated into specific specialists, but in small but growing teams they tend to fall under the manager's domain. But again, a good manager is always trying to make the system work better, so they'll often be wanting to get other people in the team to help with a lot of that too, once they have established a way of doing something. They might want the team to promote their work more too, or to keep track of customer contact, or even to come up with ideas for the business direction.

    Ideally, they're like little mini business owners. If they stop working like that, or never start, and are just performing secretarial duties of a particular kind then they're pretty much not managing.

    Should they be rewarded more than the experts they manage? Hell no (unless they really are exceptional).

    Depends on the experts, and what they themselves add as value. Some experts are, mostly because of oversupply of the expertise, not in themselves actually that valuable. It is a lot easier to find 10 good programmers than one good manager of programmers. Programmers are churned out from educational programmer factories. Yes it's quite tricky and difficult work, but it's also something our education system is geared towards producing by the tens of thousands every year. By the end of a couple of years in work, they're highly productive - the skills are all well defined and the ability to meet them can be measured (not perfectly, but at least somewhat). But good management, a much softer, less easily measured skill, that can take a very long time to master, sitting in a mission critical position in a business (since failures can impact on the performance of the entire team), can be hard to find.

    I've worked in number of places where management aren't the highest paid people, and certainly the wealthiest people I know aren't just managers (although they all have to do some management). It's not uncommon for salespeople to get extraordinary pay. Sales managers, even more so, but not because they are managers so much as because they've worked out how to do sales as a team effort. Astonishingly high pay goes to merchant bankers who are mostly just consultants. Lawyers and doctors can get huge money too. Again, those of them who manage to do it in a team tend to get more, because they've worked out how to be more productive. But a head surgeon is still a surgeon, and usually a bloody good one.

    The biggest pay of all goes to actual capitalists, though. People whose own money is at stake. To call someone who sets a business up, makes it hum, and then creams it a mere manager who should get paid less than one of the workers they gave a job to is not really seeing the job for what it is. A good many capitalists lose their shirts and face years of bankruptcy for taking the risky path that it is. In bad times, most of them go this way.

    Auckland • Since Nov 2006 • 9036 posts Report Reply

  • Hard News: Campbell interviewed, in reply to NBH,

    Management do get some unfair cops, especially when treated as a stereotype. In practice there are many different skills in management, many different positions too. Some people might make a good low level manager, but not a good middle manager. Or vice versa. Some people are good at managing only one kind of team, others can manage many kinds.

    In NZ in technical fields there is a real problem of finding people to even do the job - for all the complaint that management don't know about the coalface, there is a dearth of people at the coalface who are actually prepared to stand up and become managers.

    For good reason - it is not always easy, and it can involve significant career risks. If you are, for instance, a gun computer programmer, it could seem like a good idea to move into management, but you'll then find you have no time for programming yourself, and in a field that changes rapidly, your skills get out of date fast. Then, should you find that you don't really like management, your career has gone backwards, you are no longer as highly skilled as before.

    If you opt to stay hands on (only an option for the lowest level management), you could have extremely unrealistic expectations about the team of programmers you lead, being a gun yourself. Most people are average, that's pretty much what average means. You'll end up over-promising based on what you yourself could do, but find that 7 people aren't 7 times faster than you, and you end up having to spend long hours at the coalface as well as managing the team. The management chores will suffer, as will the programming, and you'll end up appearing to fail despite extraordinary efforts (and probably achievements too, in an objective sense). It is actually fair under such circumstance that you demand a lot more money to compensate an extremely stressful job.

    The jump from low level to middle involves giving away all true connection to the coalface. The best that can be maintained is familiarity, and the skills that are meant to be developed are pure management skills. Of course at this point there is a genuine detachment from what the actual business of the firm is. Quite often you find that you are at a real disadvantage to people who specifically trained in management and have already internalized the entire process, and only lingering kudos from colleagues continues to drive you forward, although in the process of resolving disputes between them, it's likely that the pool of people who feel any genuine affection for you has shrunk considerably, and your actual peers are beginning to be the other managers.

    Senior management is a whole 'nother thing (and middle management could have many layers to wade through first). People can actually parachute straight into it, and are quite often better for it, better never to have gone through the middle at all. Often they have come full circle, not having deep management skills at all, but much more specific technical skills relating to some aspect of management that they will do as part of the top team in the organization. They could be an accounting expert, or a fantastic salesperson. If they were once a great technician, it will have been quite a while ago, and their stories will be the tedious ones elders deliver to show their roots to the masses, their mythologies upon which they base their ongoing art. These anecdata can be as likely to hinder as to help any actual progress the firm could make. Or maybe they're like politicians, friends with everyone, great at smiling, and gone in an instant, taking care never to promise anything specific, but leave an impression of safe hands and genuine concern. Managing their own image becomes extremely important.

    Who really wants to do all this? It ends up being quite a small segment of the population, and the ones who are actually good at it are likely to be as randomly distributed as people who are good on the coalface. Most managers will be average, because that's what average means.

    Auckland • Since Nov 2006 • 9036 posts Report Reply

  • Hard News: The epitome of reason, in reply to jh,

    left-wing academics tend to hop in the lift to the top floor.

    Pretty sure Russell didn't come in the top floor of any academic process, given that he has never even been to University. AFAIK, he worked his way up through journalism the way it used to be done, from the ground floor, and if you're confusing him having decent writing skills with being an academic, that says a lot about your expectations of journalists these days.

    Auckland • Since Nov 2006 • 9036 posts Report Reply

Last ←Newer Page 1 2 3 4 5 6 904 Older→ First