[LINK] Grocery Choice - what's the problem?
Craig Sanders
cas at taz.net.au
Wed Jul 1 11:40:21 AEST 2009
On Wed, Jul 01, 2009 at 12:03:02AM +1000, Sylvano wrote:
> On Tuesday 30 June 2009, Craig Sanders wrote:
> > On Mon, Jun 29, 2009 at 09:54:38PM +1000, Sylvano wrote:
> > > We kind of don't need government, choice or the retailers to keep
> > > track of the pricing of things.
> >
> > actually, government and legislation ARE essential - without them,
> > if any service like the snaptell one you mention ever became a
> > significant factor in the marketplace, it would be astro-turfed into
> > irrelevance.
>
> I take your point, though I did only say that a user generated price
> comparison internet based utility was something to *consider*. And
> it comes off the back of the particular observation, as I mentioned,
> of a market research service that does in fact operate that gathers
> detailed product pricing info from regular people as part of a market
> research panel.
>
> And that service is an example of something that exists, which does
> not require government or legislation - or retailers - to achieve it's
> aim.
except that, ultimately, it can not achieve its aim. as soon as it comes
close to doing so, the usefulness of the site will be buried under
mountains of bogus data. real data may still be in there somewhere
but finding it will be beyond the skill, time, and patience of almost
everyone.
> The consideration of "the option of people uploading info to an online
> price sharing web site" *is* a compelling one, for that very reason
> that it's not reliant on government or retailers.
and the information uploaded is unverifiable and anecdotal. like any
anecdote, it's not worth very much. worse, nobody is accountable for
the accuracy of the information.
if a supermarket lies and says that their current price for product X
is $y, they can be held accountable for that lie.
if some anonymous person/astroturfer/bot lies that they saw product
X in their local supermarket for $z, there is no accountability. and
there is no reasonable way of distinguishing between such liars and any
genuine people who are accurately posting their observations, especially
if (when!) genuine users are outnumbered tens or hundreds to one by
astroturfing bots.
if you think that the existence of such bots is unlikely, then consider
the fact that well over 90% of all email transmitted today is spam
sent by bots. and a significant percentage of the comments on blogs
and similar sites are also spam posted by bots. if and when there is
an economic incentive, they could quite easily swamp crowd-sourced
pricewatch sites. there are tens of millions of zombie Windows machines
out there on the net, it would take only a tiny fraction of 1% of them
to completely outnumber legitimate contributors.
the corporate world has so far shown no great reluctance to use the
services of spammers - although they learnt to do it indirectly. IMO,
one of the great untold IT stories is that of the link between spam and
the junk-mortgage economic collapse of the last year. we had years of
non-stop mortgage spam before the collapse, suckering stupid people
into mortgages they couldn't afford.
> But there are issues to face, of which there was the one you raised,
> being astro-turfing. Particularly where "those with deep enough
> pockets or by skilled memetic engineers" can subvert a crowd-sourced
> info site. But I have to admit that I'm not at all clear on how
> government or legislation would helpful toward stopping that kind of
> thing. [Net Filters Stop Porn, Net Filters Stop Porn, Net Filters Stop
> Porn...]
government legislation can't stop the subversion of a crowd-sourced
site. which was (part of) my point that "crowd-sourced" sites are
pretty much useless in the long term - as soon as they get big enough
to be have an impact in the real world (rather than in the idealistic
imaginings of uncritically enthusiastic technofetishists), they're big
enough to be worth destroying by bombarding them with garbage data.
government can, however, mandate that supermarkets provide accurate data
to a specific watchdog site (such as the one that was going to be run by
Choice) - with penalties for providing false data.
which is why i said that government and legislation are essential.
and it's a perfect fit for the ACCC's existing role as competition
watchdog, whether they run the site themselves or "outsource" it to
a consumer advocacy organisation like Choice and just provide legislative
oversight and a big stick to ensure compliance with data-accuracy
requirements.
> By comparison, the actions of the users themselves to provide a
> corrective to the input of "bad data," whether in the context of
> Amazon reviews, wikipedia entries or whatever, is difficult to ignore.
it's hard enough for such sites to correct bad info from individual net
kooks. impossible to resist a concerted effort from vested interests
with money and resources behind them.
craig
--
craig sanders <cas at taz.net.au>
More information about the Link
mailing list