meta-roj

back in december, i left myself a note to do some homework and make this comparison… with the election now just five months away, i guess it’s time to get this on the record. this won’t be as fully-developed as i might’ve hoped, but that’s what comments and links are about. the comparison between slot machines and voting machines has been made before, but i wanted to tie the two companies together in a sort of risk-assesment way.

in 1990, larry volk was killed shortly before he was to give testimony on how he programmed video gaming (as in gambling) machines for american coin – programmed them to cheat. nearly a decade later, the story got hot again with the confession of his killer. the story is great stuff for crime drama and mafia stories, and that’s been covered well already, but that’s not what interests me.

the issue for me is that even with the intense scrutiny of the gaming commission and state gaming control board – through a formal and documented procedure for review and approval of gambling machines – even with all those measures in place, bad code managed to get into the machines, get into the field and rip off players.

i’m not terribly concerned about ripping off gamblers, at least not personally. part of me wants to put some distance in this, and say that rigged gambling machines is, to some extent, part of the gamble. i guess the problem with that attitude is that it ignores the assurances that a gaming control board is supposed to provide. slot machines aren’t supposed to be part of the whole wild west “you takes your chances” mentality anymore. you’re supposed to know the gamble you’re taking, and i can appreciate that. but, the stakes are personal, and when we’re talking about gambling machines, the bets generally aren’t too huge, so the failure modes in this environment don’t get me too uncomfortable. ok, so you lost a dollar. ok, so you lost ten dollars. might be a big deal to you, but i’m sure i’ve been scammed out of ten dollars many times without even playing the machines.

the precise nature of the hack that is at the core of the larry volk/american coin story isn’t really important. what’s important is that there is a layer of protection between the player and the manufacturer, and that layer of protection is elaborate, well-funded, staffed by professionals, and does full reviews of both the hardware and software both before and after machines are approved for use in nevada. despite all that, about 1000 american coin machines were approved and in use for some time before being siezed by the gaming control board in july 1989.

of course, for-profit gaming is a bit different than civic-duty elections, and there are practical concerns – inspection and monitoring of slot machines in nevada has to be cost-effective, and there are tradeoffs made under that burden. machines are spot-checked (as opposed to individually checked) and they are only checked every couple years or so.

i’ve been on the diebold case for some time (watch your touchscreen, diebold compromises everything and takes cowards down with them, the flipside of diebold at swarthmore, diebold on election day, some diebold news, election day trouble, diebold woes in california). the latest that i mentioned was about a revised ethics policy at diebold, which they, presumably, instated to address some concerns of the lowly voters and their advocates. this isn’t just about diebold, of course, but diebold is here on the blog with a developed history.

if nevada can miss a few guys within a company chipping machines, what chance do our elections have with underskilled legislators and intermittantly-constituted election review boards have? are we supposed to rely on an ethics policy?

for elections, the stakes are a lot higher. vote on paper. be counted. be recountable.

update (2004.12.02): this post recently made an appearance over at democratic underground, but with a truncated [bad] link. i’ve fixed that on my end, but it would be better for them to fix it on their end. and if anyone is interested in working on democracy and technology, i’ve got a little project cooking in the background here, so get in touch.

i’ve spent a considerable amount of my blog-space commenting on how things should not be done, and i guess that trend will continue today… of course, i have all the answers, that’s part of the geek paradigm, but what’s the fun in letting it all out of the box at once?

this began with a speech given by carly fiorina at the university of maryland (you can skip the rah-rah bits about the company).

carly fiorina, cio forum / inforum 2003, robert h. smith school of business, university of maryland, october 10, 2003

It was a great company. But in many ways, it had fallen so in love with its legacy and its past that it had forgotten to build its future.

as i started making notes, i ran into a piece on the “15% delusion” by carol loomis. it appeared in fortune magazine back in february of 2001, with the wounds of the burst bubble still fresh in many minds and mutual fund accounts. unfortunately, fortune only leaves the first few paragraphs of their archives available, so there’s no real point in linking to that.

the gist of the article, if i may be so bold as to boil it down to a sentence or two is this: don’t make predictions you know you can’t keep. and if you do, don’t keep doing it.

the context is the business environment – the big huge public companies, the big huge brokerage firms, and the constant barrage of data and news and opinions that drive the markets (that is, stock markets) up and down and sideways from second to second. it’s about the pressure to perform… no, actually, it’s about the pressure to out-perform, and it’s the same thing we see in so many interesting places in this society these days. it’s not good enough to be good enough anymore – you have to be better. you have set high (that is, unreasonable) expectations and then you have to exceed them (that is, impossible) or you are punished.

loomis mentions the 15% growth target, but her article is buried. fortunately (ahem), the guys over at the fool picked up on it and ran in their own direction with the 15% fallacy. that one’s linkable, so let’s go there for this adventure. i’ll leave the math homework to the fools. their perspective (investor perspective) is useful to explain the pressures on management, which is where i really want to go with this. investors want growth stocks, growth stocks grow at 15% per year, and that’s five times the historical average economic growth in the united states.

the problem is that not every company can outperform the average (i leave the math to back up that assertion to the reader). even if you’re in the company that is growing way beyond average, and for a good reason (perhaps a new company, or a new product, or something else meaningful), eventually with 15% compounded returns, that one company would suck up the entire market for everything, and that’s not a good thing. but in order to keep the stock price up, you have to promise those better-than-average returns, until the rug gets ripped out from under you and you either lower expectations or miss expectations, and the bottom falls out of your stock. and there’s the damage at the highest level – between the company and its investors.

so there’s a massive incentive to take the short-term gain at the expense of long-term results and to cheat – to fudge the numbers as much as possible for as long as possible and get out while the getting is good. hit those 15% targets long enough, and the stock price will grow even faster than 15% – for a while. then pile all the damage into one quarter (if possible) suck it up, run it through bankruptcy (if it’s that bad) and who really cares what comes out on the other side? there’s the real damage – the company rots from within as decisions accumulate that “work for now” but eventually expand like a cancer.

a favorite whipping-boy industry for this has got to be telecommunications. the problem is that once you’ve got a chunk of wire in the ground, there are only so many ways you can earn money with it, and several of those ways are disappearing with new technologies. there used to be a pretty solid bet that you could hook someone up to the wire, and keep them (or someone that moved into that house) hooked up to that wire indefinitely. all things being equal, just raise the price at 15% a year and you’re golden without even hiring enron’s accounting team. but all things are not equal, and now your wire has to compete with cell service and satellites and voip that’s going through the air, or down someone else’s chunk of wire. and the consumer is demanding things that your wire can’t physically deliver.

eventually, you get worldcom. to meet growth expectations, it had to suck up every company it could possibly leverage to buy, and when the bottom fell out, it found itself not only missing expectations, but leveraged so far out over the cliff that there was nothing to do but wait for the coyote to look down. depending on how you do the accounting (and that’s another issue), worldcom just made up something like $12 billion along the way – which promptly vanished, and then some.

the problem with this (and i’m sure i’ll catch some dissent on this point) is that it builds companies on these elaborate, diligent, fully-developed and amazing frameworks of vapor. it works (in as much as it does work) for a while because everyone who is in the game sucks up the same vapors. but with these goals, targets and short-term plans, these companies with all the accumulated resources spend so much energy planning to get bigger that they miss the opportunities to actually do something about it. the opportunities happen, randomly, but if they don’t fit the company plans or the division plans, or the preconceptions of the person who’s promised 15% growth in something, then they just slip right by. worse, the oppotunities are cannibalized for short-term returns to meet the next reporting period promises, rather than focusing the available resources on longer-term prospects.

a couple years after the fool picked up on this subject, jack trout came up with another version of the same story, which can be found in the economic times of india. he’s even got a nod in there back to loomis’ article, and a small bit of it got picked up by a blogger:

Why big brands fail [ryan’s hope]

I agree that growth (through innovation) often happens in spite of goals and forecasts, not because of them.

i hope i have the wisdom to assemble brilliant people that won’t forget to build the future with me. in the meantime, please don’t forget to put a cover sheet on those tps reports.

so, who wants to take me to lunch with warren?

i’ll bring the heckle…

we should also bring bill moyers and walter williams.

reported in the washington post, halliburton (which we should all be familiar with by now) has fired two consultants in connection with “violations of halliburton’s code of business conduct.”

one of these consultants is a. jack stanley – picked by cheney as chairman of the kellogg, brown and root (kbr) subsidiary of halliburton. the other is william chaudan

Halliburton Fires 2 Consultants [washington post (key)]

“This news is disappointing, ” said Wendy Hall, a Halliburton spokeswoman. “But it’s not representative of how Halliburton or KBR does business.”

right. so far the business model at halliburton seems to be overcharging clients like the us government. kickbacks and bribery seem to be more badder and so off-limits. there’s always the chance that the bribe could come to light, and overcharges can simply be attributed to clerical errors. i mean, even the client makes mistakes with counting.

it’s a rational approach – if you get busted for the overcharges, just pay whatever it takes to make it go away. if you don’t get busted, you keep the money. if you get busted with bribe money, that’s going to set off a few more alarm bells. it is, after all, a violation of the foreign corrupt practices act.

revisiting a brief discussion we had several months ago, frequent meta-roj blogologue contributor barry ritholtz is fed up, and he’s not going to browse there anymore.

it’s a thread that’s been woven through the meta-roj blog… sometimes we bounce it off the television, and sometimes the recording industry and often woven tightly with the theme of active vs. passive “consumption” (and it’s always related to spam).

barry’s got a website that he visits, regularly, but has apparently become increasingly annoying. today, the relationship is over, and the site in question is going to have to either a) live with losing barry, or b) jump through flaming hoops to win him back. i’m betting on the former. i’m also betting that they have about the same attitude as sunncomm – “We hear from less than half of one percent of people…” – and they’re missing the big picture (to borrow a phrase).

to succeed in this vast universe of options, you have to be useful without being annoying. google’s done a good job with that, and despite their occasional flirtation with evil, they do self-correct rapidly.

i’ve got my own annoying-google story from sometime around 2000 – my habit then (before the advent of the tabbed browser) was to leave several browser windows in the background on the desktop available for quick access – and google was an obvious choice. then google decided to add a little bit of code to their page that would a) refresh the page every few minutes and b) bring the page “to the top.” you can imagine this was excessively annoying to have the google search page pop forward at the most inopportune times, like when i’m writing something in another application and end up searching for “oice).” anyway, i complained (i was one of the tiny percent that they heard from, i didn’t want them to take my google away) and i’m sure several other people did too. i’m equally sure that many people did not. google listened to the tiny percent, and the annoyance vanished within days.

true geeks among us might try to make the case that barry’s problem is with internet explorer and not with the site in question – and they might be right. it’s entirely possible that barry’s better approach may be to switch browsers and not websites, but the site still got the blame. nobody (not even barry) should have to jump through hoops (even simple hoops) to keep a company from annoying them, and companies better find that religion.

advantage: customer.

24hdc, the dotcom startup sold on ebay closed earlier at US$2, 026.00.

I seem to recall reading somewhere that they needed $1000 to break even, so congratulations on the profitable adventure.

long, long, ago… it seems when i was putting more thoughtful pieces together in this space, i bounced off transparency as a business advantage, and did so in a typically oblique sort of way.

i’m a fan of transparency, in general, and the story about vw’s phaeton factory [car and driver] provided a tangent, but i never really got back to it as a thinking piece. this probably won’t be either, i just want to move the thread up many months, with a nod to fred, who left me this hook to hang myself on….

i’ve explored the problems with lack of transparency here in some gory detail – mostly related to diebold and the electronicification of elections (theme song: elections should be transparent) and government, most particularly the department of justice, under the fearless and righteous leadership of john ashcroft. and that brings me full-circle, since i stumbled into fred’s blog via jon stewart and john ashcroft, and i was so moved, i even left a comment.

so, fred is a vc, and i try to be a positive kinda guy in general, so this post may not have a lot of original thought in it, but i’d like to wrap up on a positive note, and remind everyone that transparency is working (or so it seems) for magnatune, and i did just dig into that again.

barry, fred… fred, barry. i think you guys are neighbors.

i won’t be able to follow this very closely (busy day here at the rojidojo), but you might… less than 8 hours to go for 24hdc.

[via suw]

primus [telecommunications] has jumped into voip to compete with vonage and at&t – offering $20/month unlimited calling.

no word on the big brown beaver.

ap reports [via the guardian] that the diebold board of directors has amended their ethics policy so that top diebold officials “may not make contributions to, directly or indirectly, any political candidate, party, election issue or cause, or participate in any political activities, except for voting”

it’s not clear if providing “badly designed” voting machines and compromising the integrity of elections in the united states is a “political activity” or just “business.”