Subcultural engagement for the 21st Century...
Barbelith is a new kind of community (find out more)...
You can login or register.

Underground > Laboratory >

New Developments in Computer Networking

16:47 / 18.09.06

Nature reports on an open-source textbook initiative.

There are millions of university teachers around the world and tens of millions of students, whose knowledge could be put to greater use, says project instigator Rick Watson at the University of Georgia in Athens. Countless essays and assignments are currently consigned to the waste bin. "It's an untapped intellectual resource."

...Anyone will be able to contribute to the new textbooks, true — but unlike wikipedia, the online, user-made encyclopedia, only an editor will be able to approve contributions. Otherwise the texts risk being wrong, long and hard to follow....

The particular goal of this project is to create free books for those students in developing countries who cannot afford traditional textbooks, which can cost $100 or more. Most current textbooks cannot be freely scanned on to the web because they are protected by copyright. And in fast-moving fields such as computer science, a printed textbook quickly falls out of date.

Combine this with the <$100 Laptop, and I think you've got the recipe for a free global university. Which I think will be an awesome thing to behold.

14:55 / 17.10.06

There's a group planning a "fork" of wikipedia, called the Citizendium.

Basically, according to their fundamentals page, it seems like they want a Wikipedia with greater accountability -- all users are identified by real name, mechanisms are put in place to prevent libel, and articles are given an authoritative or non-authoritative status (I think this means some kind of review process by editors, who seem to be "super-users" or experts in a given field).

Does anybody know more about this?

Think it'll get somewhere? It certainly seems more academia-friendly and authoritative -- but that also makes it seem to run counter to some of Wikipedia's ideals.

Good Intentions

02:02 / 26.10.06

Wikipedia is currently the best thing on the internet. But it won't be able to take the steps to being the fulfillment of the Encyclopaedicians' dream until a few things change, and I don't think enforced accountability is the big one. I think what has to happen is that those involved need to have a closer interest in the project than seeing good info online (that small incentive has already made Wikipedia the great source that it is). Because the current interest the contributors have is so comparatively small, the belligerence of 17,000 teenage Randian twits overwhelms whatever will to contribute (and live) the truly informed have, because FreedomFromSheeple14 really wants the page to say how awesome the free market is.

Kiltartan Cross

19:01 / 28.10.06

An idea strikes me - currently, we (as a species) do a lot of our heating - of houses &c. - by directly converting electricity into heat.

It would be possible to add an intermediary stage - a computational one. We do this a lot at present, of course; the energy used to power your PC will ultimately end up as heat. Each PC is, in effect, an, oh, let's call it 250W electric heater.

But suppose we used computers instead of our heaters? Instead of having an electric fire on the wall, imagine a bunch of processors, some rudimentary architecture (no video or human-computer interface, a small amount of local storage), and a network connection, all bundled up into a small box. Allow that box to connect to things like, say, the SETI at home stuff, or the Beeb's climate change predictor thing - any of the big, multi-computer networking projects - and we'd get both surplus computational power and heat instead of just wasting the energy direct to heat.

Drawbacks as far as I can see would be: you'd need some way of connecting each box to the home's internet connection. You'd need to do some work to make sure the major projects could talk to each box. You'd have security issues ensuring that only "legit" projects could control the boxes. Initial cost would be higher than a wasteful electric heater (although, not, perhaps, hugely so, especially if they were being mass-produced).

Chip-manufacturers get a big win, spare computing capacity goes through the roof, and the world, if not exactly saved, isn't particularly worse off than it will be anyway.

Anyone got any thoughts, plusses I haven't seen or more crippling obstacles?

(or a link to whichever bastard has got there first!

)

19:04 / 28.10.06

I think the most obvious problem with that proposal is that computers are actually quite inefficient producers of heat for the amount of energy put into them, and also consume an enormous amount of energy in the mining, design, programming, synthesising, fabrication, packaging and delivery of their chips, fans, operating systems, boxes, motherboards and the like - often causing many many larger ecological problems including the distribution of heavy metals.

19:12 / 28.10.06

More generically I'm interested in the Citizendium proposal, because it seems to me that there is clearly a balancing act that has to be undertaken between ease of contribution (and hence growth of the encyclopaedia) and the trust that you can place in the entries on the page. If - for example - you examine the history page for Yahoo! on Wikipedia, and you start going through and seeing the differences between the relevant versions, then you fairly quickly start to see that these pages experience an enormous amount of vandalism every day, which - for the most part, but not completely - gets cleaned up each day.

More troubling is the stuff concerned with whitewashing or manipulating your public representation on the site. It's surprisingly simple if you're a medium well-known person with an unflattering representation on the site to massage it into one that is less unflattering (I've seen people do this) but still appears to be roughly plausible and superficially 'fair'. There are a lot of problems with this stuff generally.

My suspicion is that for this to work, you have to start off with a corpus and a user-base as large as Wikipedia itself and you have to start working out how to add a little more structure and rigour into the maintence of the site. You can't go off and start your own. But it's going to be profoundly difficult. There's a fascinating article by a precocious guy called Aaron Swartz in which he confounds the myth that most of Wikipedia is written by a few hundred people, suggesting that while a few hundred (maybe thousand) people do the bulk of the editing, the actual bulk of the content is contributed by a much much larger group of people, who often only contribute once when they find a page that appears to be broken or have inadequate information in one or more areas. They come in, add that content and leave. Under the Citizendium model, that couldn't happen, so you have to wonder whether it has legs.

I'm interested - more generally - do people here use Wikipedia a lot? Do you trust it? Do you have a sense that it's getting better? Or getting worse as it hits the mainstream?

20:02 / 28.10.06

I like Wikipedia and have edited a few pages there on and off. Mostly, I suppose I'm like a hybrid between the content provider and the editor/community member -- I tend to contribute only on a handful of pages that are about things that interest me, although I do tend to fix spelling errors & word choices on random pages when I see 'em (which isn't often).

I actually think Wikipedia is lurching towards incomprehensibility in a few areas -- it's attracting a level of expertise in various fields that makes certain entries largely incoherent to outsiders. I'm not sure this is entirely a bad thing; the same criticism could be leveled at the Britannica for some things. But it's something I've noticed that other people don't seem to talk about.

Kiltartan Cross

23:03 / 28.10.06

I think the most obvious problem with that proposal is that computers are actually quite inefficient producers of heat for the amount of energy put into them.

I think I know what you mean, manufacturing costs included - but in general running, they're (close to; presumably there's RF emission of some degree) 100% efficient, no? An electric fire could be considered 100% inefficient, in that no useful work, other than the production of heat, which isn't so much a product as an end state, is done.

I dunno. I wonder what the effect of the great inroads made by electrical appliances of all kinds (from lightbulbs through fridges through mobiles through TVs through (...) ) has been on general consumption of energy for heating?

At a rough estimate we're looking at, what, maybe 2kw for a few computers, a TV, and a bunch of lightbulbs switched on? So we're already at a state where an average house has the equivalent of a decent electric fire running constantly; not to be sniffed at. Multiply that by, say, ten-twenty million houses (for the UK) and it's a hell of a lot of heat.

---

But ach, yeah, on Wikipedia, I don't think that a trend towards greatly specialised / generally incomprehensible articles is in any way a bad thing, so long as subject articles at least start in a "layperson's" format, or so long as there is a general "covering" article. The more the merrier, really. It's sobering, in a way; it's not hard to come away dazed from an encounter with the collected brains of humanity, but that's a good thing, right?