Internet Alchemy

Well, there are signs of progress in the RSS world. Dave Winer has offered
an olive branch to the
RSS 1.0 group.

The goal is to clear up confusion in what RSS means. I make
this offer to show that we’re willing to share the pain to help RSS
re-establish a clear identity.

Essentially Dave is proposing that both forks of RSS (RSS 0.92/0.93 and RSS 1.0) are renamed and that
the two groups go their separate ways, one as a simple headline syndication format, the other as
a extendable metadata format.

Reaction on the syndication and rss-dev groups has been mixed, but mostly positive. Dan Brickley expressed his concerns in
a posting to the rss-dev list:

We know from the various XSLT experiments that we can
inter-convert between RSS flavours, that 0.93 extensions can be shadowed
by RDF properties, that we can deploy XHTML (eg. the W3C home page) and
use XSLT to aggregate RSS direct from the XHTML doc. I fear that
creating new names for existing and future technology will simply add
more confusion to the mix.

My feeling is that we should look on Dave’s offer as a positive gesture and we shouldn’t
ignore it. Too much time has been lost and too much confusion sown over the past twelve months.
If you’re involved in RSS development in any way, please take time to vote in the
rss-dev poll and express
your opinion. Should we accept the gesture and move on or stay as we are, slowing progress?

This piece is the first part of a series chronicling the construction
of a 20 mile wireless lan by the O’Reilly network
team: A Wireless Long Shot

Now that we’ve achieved Ethernet-like speeds over a 5-mile wide valley, and
simulated a 20+-mile link over the same, we’re preparing for our 21-mile shot.
Our experiment proved that it is theoretically possible to drive 802.11b signals
well over 20 miles, using stock equipment.

This has to be the strangest thing. Someone has stolen a copy of this site
and has put it on a server being used for portscanning attacks. I was mailed earlier today by a sysadmin who
asked if I knew anything about the attacks. I was confused because I couldn’t understand how he’d associated
me with that IP address. When he mailed back and suggested I visit that address in a web browser I was gobsmacked
to see an old copy of Internet Alchemy up there! Not only is this person portscanning random networks but they’re
masquerading as me for some reason. Very very strange….

It’s nice to see the old AltaVista interface back. For a long time AV was my search engine of choice, but
the past few years of excessive advertising and pushing of features most people have already really turned
me off. Pandia have a nice article
discussing some of the changes happening at AltaVista.

Is Content Syndication A Viable Revenue Stream?

The other benefit of syndication, many analysts believe, is increased traffic.
But Mardle cautions against using such metrics. “We have seen the idiocy of
“quickly building traffic” as a business objective. That is what killed most
of the dotcoms. Traffic doesn’t mean squat, it is not the same as audience.
Got to get that across,” he says.

Amusingly I found the above article while visiting Figby a
news aggregation site. The article appeared in the right hand column three times from three different
sources which just underlines the importance of the following excerpt:

“The syndication method is an interesting one because it presumes
that there is no readership overlap to devalue the information,”
Mardle adds. “However, the Internet is an excellent multi- channel
delivery system and I will quickly discover that the information has
massive overlap because it is syndicated, which will naturally devalue
all the places on which the information is repeated.” If a consumer
is checking six or ten sites a day, he suggests, they would expect a
wider range of information or they would be unlikely to return.

Larry Wall has published
the second of his Perl 6 apocalypse articles.
With this article we start to get an inkling into just how different Perl is going to be next time round…

…Perl programmers must learn to write @foo[1] where they used to write $foo[1].
I think most Perl 5 people will be able to get used to this, since many of them found the
current syntax a bit weird in the first place.

Various special punctuation variables are gone in Perl 6, including all the deprecated ones.
…
$#foo is gone. If you want the final subscript of an array, and [-1] isn’t good
enough, use @foo.end instead.

Typeglobs are gone. Instead, you can get at a variable object through the symbol table hashes that are
structured much like Perl 5’s. The variable object for $MyPackage::foo is stored in: %MyPackage::{’$foo’}

Leigh Dodds has published his new Schematron paper
from XSLT UK - Schematron:
validating XML using XSLT

One person’s nightmare interview as Napster:

One friend of ours told me about a recent experience he had at Napster. They
were apparently hiring QA people to “develop automation” and he thought,
yeah, I can do that. Talked to a recruiter on the phone. They asked what he
was looking for, salarywise, and he threw $100K out, figuring they ought to
be able to afford it. They didn’t even blink so he went in for an official
interview.

At the first interview, the interviewer mentioned at one point, “You’re okay
with putting in long hours, right? ‘Cause we tend to put in a LOT of hours.”
He thinks, yeah, okay, some twelve hours days, 60 hours a week, I have no
personal life, I can do that, sure.

At the second interview, the interviewer says, “Did the last person tell you
about the long hours? Whatever she said, she was probably understating.” It
turns out this second interviewer, who would have been a direct peer of his,
had come to the interview at the tail end of 36 straight hours at work. She
was going to go home for twelve hours and then be back to do another one.

Then she showed him what their “QA department” is doing these days.

They don’t even have cube farms. They have long tables set up with systems
in a row at which people are sitting around the clock, manually looking
through endless lists of every single file that goes into and out of napster,
searching for copyright violation material concealed in every possible way.
M3TALL1CA PHAD3 2 BLAKK, ding! Flag it. KCALB OT EDAF ACILLATEM, ding! Flag
it. Anything that sits in their hands for 72 hours and turns out to be a
copyright violation, they’re fucked. So they have rows of drones looking at
everything and trying to figure out every possible way people can disguise
band and song names.

“We need to develop a better way to do this. You up to the task?”

He got the fuck out of there.