Monday, February 14, 2005

The next generation of computing? Here are two articles from the Economist, the first is about the new cell chip, and the second is about a new, faster access form of data storage. I can't help thinking that the combination of these two technologies will lead to the next surge in multimedia evolution. It makes me ponder the possibility that, in terms of the technology of digital information, we are on a par with the early commoditisation of electricity: it's widely adopted, but non-standard, barely reliable and barely understood. Yet the possibilities will extend far beyond the pioneers' imaginings.

Thursday, February 10, 2005

Tablet flaw hard to swallow (ZDNet UK News)... basically using the pen on any tablet PC leaks memory, so every day your memory fills up forcing you to reboot. Every day. This is a product that's worth at least $1000. In any other industry such a flaw would be grounds for a recall, compensation, lawsuits etc. But in software, it's a corporate 'oops! Please hold while we figure out how to fix it'.
Why do we tolerate this? How long must we tolerate this for?



Surely the strongest argument for open source is that it is ethically more balanced with the standard EULA offered by most 'professional' software. Currently, when you buy most software, you indemnify the vendor from just about any fault with their software. And if you want fault resolution, you pay extra for it. If they won't take responsibility for their software, why should they be entitled ownership? In open source, the positions are clear and more even-handed: no ownership, no responsibility - just pay for the service.

How-To: Podcasting (aka How to get Podcasts and also make your own) (Engadget)... well, it seems to be a how-to day today. Just need to find a decent podcast search tool... and maybe try to persuade the BBC to adopt it for their radio shows - subscription/license fee only, of course.
Two great thinkers of our time discuss the profound (Neal Stephenson's Past, Present and Future) and the bizarre (William Gibson's blog) follow the link in the birds and bees entry in the latter.
Decent article on secure connections on Mac OS X -- here's a useful, comprehensive blog article about using secure sockets (SSH) in OSX. It explains the rationale for using it, as well as the techie stuff in layman's terms. Also recommends a handy config product: SSH Tunnel Manager.

Monday, February 07, 2005

Slashdot | Grand Unified Theory of SIMD. Wha...? Well, here's one for those who think that graphics people only prefer macs because they look nicer. Single Instruction, Multiple Data (SIMD) is an essential part of graphic manipulation, and is yet another one of the hidden wonders of Apple's OSX. Have a look at the comments below the article for explanations and other good links.



Also, the cell chip, mentioned in the article, is due for launch today. Touted for the PS3, it is supposed to be supercomputer fast (16 teraflops in a blade is one quote I've read). Wonder if it will fit into a Powerbook...?
Economist.com | The economics of sharing. This is one subject area I've been pondering for a while: how can open-source development work, economically, and can it be applied in other areas/industries? As ever, the Economist does the thinking for you, while at the same time provoking more thought. The essence of the first part of the question is the "non-rival" nature of such goods (your use doesn't interfere with my use) and their network effect benefits (the more use, the better).



The second part is harder because there is another key to open-source sharing: the cost of distribution, which is negligible. Clearly, for physical goods there will always be a non-trivial cost component. However, there are still plenty of semi-virtual goods out there that could work, including radio frequencies, computer processing and bandwidth (eg. SETI).



Key quote in all of this: “Social sharing [represents] a third mode of organising economic production, alongside markets and the state.”

Economist.com | Computer circuits: Barred and latched. Here we have it: the missing piece of nano-circuitry. They had already sussed the basic gates (AND, OR etc.) on the nanoscale, but not transistors. Enter the new nano-transistor, a breakthrough that some electronics genius with no sense of marketing decided to call the "crossbar latch". Great stuff, can't wait for my first Crossbar Latch Radio..!
... and so it came to pass that Microsoft was based on unix all along. While this isn't exactly news, it is a useful reminder of the reality behind Microsoft's version of history. They started out as opportunistic as some of their latter day competitors, so their implication that the fragmentation of unix was a bad thing, not to be repeated, and implying that they are merely custodians of some OS standard that must be preserved is rich, to say the least. As is Gate's comment about AT&T not realising what they had (in unix) and be very hard to deal with. Sound familiar? Although I think this time round it is worse: the protagonist knows exactly what it has, and is willing to fight, fair or foul, to keep it... even to the extent that it has created a nemesis that is essentially a guerilla movement: amateur idealists giving their software away for free: Linux. And what is this software based on...?



UPDATE: just saw this (ZDNet) containing an analyst's response to Bill Gates' complete misunderstanding of open source. He cites another interesting reason for not having interoperability problems: less incentive to fork.

Friday, February 04, 2005

The Story of the PING Program In many ways, this is open source before open source, in that it's one guy with an idea who sits down one evening and changes the world in a small but significant way. That, to me, is the real revolution of the internet. Sure, you can buy stuff online, you can download stuff (pirated or otherwise), you can even make cheap calls, but the crux of the medium is the sharing of ideas. It is an incredible, unfathomable medium for ideas, be they bits of code, songs, films, discussions, or documents of trade, business or research.


Another fine example of this is Bram Cohen's BitTorrent, which is nicely covered in an article here.