[Thu Aug 31 15:37:19 CEST 2006]
OS Weekly published today a short piece about the 2.6
Linux kernel that sort of confuses me. I just don't understand why they
didn't publish it months ago, especially since it doesn't truly provide any
great insight. In any case, the following snippet is worth transcribing
here, since it's the sort of information users ask about every now and then,
and it's quite difficult to find when one needs it:
In Linux 2.6, the number of possible users has gone up to 4 million from
65,000. This is achieved by increasing the addressing space from 16-bit to
32-bits. The number of PIDs (...) has gone up from the existing 32,000 to
a whooping million.
Just trivia, but it's not always easy to find when one needs it.
{
link to this story}
[Mon Aug 28 10:56:17 CEST 2006]
Here we go with one of my pet peeves again. Yesterday, my son needed to
download a few pictures that he had taken on his Bushnell binoculars, which allows him to take
pictures with an embedded digital camera and download them to a computer via
USB. As it tends to be the case with these products, the manufacturer only
cares about Windows, and it doesn't even bother to release any drivers for
Apple's MacOS, Linux or anything else. The end result is that I was unable
to download the pictures my son had taken into my laptop running Ubuntu (gThumb simply crashes) or my wife's G4 running MacOS X. I have no choice but to boot my laptop into Windows 2000, if only to download the
pictures so I can then move them to one of the other computers in the home
network. So, did it work? Sure it did, but that's not the reason why I'm
writing these lines though. As it turned out, a few days ago I had bought a
wireless
optical keyboard and mouse set manufactured by Trust (yeah, I know, it's
cheap... whatever!) that plugged into my computer via USB. Well, all I had
to do to get it working with my Linux system was boot up the machine and
that's all. Now, how about Windows? Well, I couldn't use neither the
keyboard nor the mouse to log in, since the login window didn't recognize
either. Even better, once I logged in, Windows recognized the mouse but
not the keyboard. I know, I know, I should install the software that came
with the product so it works fine under Windows. That's fine, but my
question then is: how could anybody maintain that Windows is so much more
user-friendly than Linux if the end-user has to install additional software
just to get the wireless optical keyboard and mouse to work when Linux does
it out of the box and withouot a need for additional software? Remember,
this is a computer illiterate don't-give-me-stinking-drivers-to-install
user we're talking about here, right? After all, that's always supposed to
be the average type of user when it comes to talking about how difficult
Linux still is. To tell you the truth, both MacOS and Linux appear to work
much better out of the box with most pieces of hardware than Windows, which
always needs us to install the damn drivers.
{link to this story}
[Fri Aug 25 17:24:50 CEST 2006]
I absolutely love GNU
screen, but there's a little problem when it comes to user documentation:
the man page is quite limited, and most of the articles and tutorials one
finds on the Net are too basic. Don't get me wrong. They're fine as an
introduction to the application, but they all appear to include just the
basics to get you going. As soon as you try to take one step beyond that,
you're on your own. To make matters worse, tha name of the application itself
doesn't help when it's time to search around for documentation. I don't know
about you, but I tend to run into a lot of hits about displays, monitors and
the like, which makes perfect sense. So, I truly dreaded searching for help
when I had a brain fart today, and couldn't remember the key combo to change
the titles. To my surprise, I found a nice document to add to the introduction for beginners
written by jeduthun and published on the Kuroshin website sometime
ago. I'm talking about the Screen's User Manual published by Delorie Software.
I imagine they simply took it from GNU, but
for whatever reason I couldn't find it anywhere else.
{link to this story}
Mon Aug 21 09:41:10 CEST 2006
Red Herring publishes a very short interview with Linus
Torvalds. I love the guy. It's not only his technical knowledge that
one has to like, but also his down-to-earth attitude, his humility, and his
pragmatism. Here's en example:
— What do you think is in store for Linux over the next five years?
— You're asking the wrong person. I have very consciously tried to
just concentrate on the technical side, partly because that's where my
interest is, and partly because I think the rest will follow. I don't think
five-year planned economies work, and I don't think it works when you do
software design, either. Linux development has always been a kind of
open market, where the development directions gets set by customer demand,
together with obviously a lot of what I simply call good taste —the
avoidance of things that are obviously going to be problematic in the long
run.
{
link to this story}
[Wed Aug 9 04:58:35 CDT 2006]
I mentioned yesterday a story about the role of binary drivers in a free OS.
Well, today I came across an
interesting, although quite old, post by Arjan van de Ven detailing what he
refers to as "a doomsday scenario" (i.e., a what-if depiction of what
the Linux world might be like if binary drivers were allowed and even
promoted. I think the entry for July 1st sums up his opinion pretty well:
July 1st. It's increasingly hard to run linux without binary modules on
most new consumer PCs. While a year earlier people would have to give up
3D acceleration for this often, now even 2D doesn't work without binary
drivers, nor does networking (both fixed or wireless) or sound. For half
the machines there is not enough linux support available at all, while 20%
use ndisweapper like
translation layers to run the Windows sound and networking drivers. The
Debian project, unable to run on most
machines now, is losing massive amount of users to Ubuntu and Ubuntu-Debian hybrids. Debian-legal and various other
project lists are impossible to read by people not interested in this
particular flame-topic. Most of the vendors who kept their open source
drivers at least somewhat updated have basically stopped doing so.
A doomsday scenario indeed!
{
link to this story}
[Tue Aug 8 18:17:11 CEST 2006]
While reading an article on the role of binary drivers in a free OS, I come
across the following thoughts regarding what the author considers perhaps the
main reason why certain vendors only release binary drivers:
The specifications don't exist. While not commonly used as an excuse, this
is the most probable reason for hardware manufacturers not releasing
specifications for their devics. The driver team within the company may well
be operating from incomplete specs, or even from something a bit closer to the
implementation.
If there is no specification of release quality, the only way to release one
is to write it. This requirement can be costly and, since it doesn't directly
lead to greater profits, it might be difficult to persuade a company's
management that writing specs is in their best interest.
The existence of good hardware documentation for the driver team, internal or
external, usually results in a better quality driver. If lack of documentation
is the reason why a company cannot release a detailed specification, some
caution is probably advisable when considering relying on that manufacturer's
driver.
Even more interesting are the following words about yet another reason not to
release the drivers to the world as opensource: the hardware is just plain bad.
A good example of this problem is the Raltek 8139 Ethernet controller chip. The OpenBSD driver for this chip begins with a comment
containing choice phrases such as "I'm inclined to blame this on crummy
design/construction on the part of Realtek..." The FreeBSD driver is even more
scathing. The file begins this way:
The Realtek 8139 PCI NIC redefines the meaning of "low end". This is probably
the worst PCI Ethernet controller ever made, with the possible exception of
the FEAST chip made by SMC.
Has this bad press affected the chip's popularity? Not in the slightest.
Almost everyone I know owns at least one device containing this chip. In
spite of the poor design, these chips are very popular among Free Software
users due to the low price and the fact that the chips work with the user's
OS of choice (even if the chips don't work as well as a more expensive
version).
If anyone still needed a reason as to why it's not always the best products
that win the competition in the free market, here's a good example. The
world of technology is choke full of cheap trash that sells in the millions
and excellent products that never made it. Yes, the market decides who is
the winner and who is the loser, but quality may not have anything to do with
it at all.
{
link to this story}
[Tue Aug 8 17:49:28 CEST 2006]
Yet another nice looking app for Linux. I sort of like Apple's Garage Band, and was wondering if there is
a similar application for Linux. Well, Jokosher seems to fit. I still have to give t a try, but reading the
description and the list of features
it looks precisely like what I was looking for. Now I just wonder how stable
it is.
{link to this story}
[Tue Aug 8 16:50:44 CEST 2006]
Linux.com
has a piece on WiFi Radar, a little Linux app to detect wireless networks
in your area that truly works. I just gave it a try (gotta love Ubuntu's ease of software management!), and discovered
a total of four wireless networks around my home here in Seville. To make
matters worse, one of them (belonging to a home) appeared to be wide open. The
user had never changed the default settings. Go figure.
{link to this story}
[Tue Aug 8 10:49:19 CEST 2006]
Ever since Linux and open source software showed up in the radar of mainstream
media back in 1998 or so, there has been a lot of talk about whether or not
the open source paradigm is, in and by itself,somehow communistic or, at the
very least, non-capitalistic. I seem to remember it was actually some top
executive from Microsoft who first
came up with this idea (was it, perhaps, Steve Ballmer himself?), but I'm not
sure, to be honest. In any case, today I read an article by Terry Hancock published by Free
Software Magazine that takes a different and interesting approach in all
this "free software is communism" debate for he argues that perhaps,
after all, free software truly is somehow communistic, although of a form that
is not related at al to the old Soviet Union and its satellites.
... the point of communism, as communists would describe it, is to
implement the communist ideal:
From each according to his abilities, to each according to his needs
In this particular sense, free software development does indeed follow. Since
there is often no direct and exclusive material remuneration for the work done,
the principle reasons for doing software work on free community projects are
the abilities and desires of the developers. We have an "interest-ocracy", as
some have described it —those who care enough about the design to do the
work, get to determine how it is done.
Likewise, the nature of software as information and the resulting near zero
cost of replication of he work leads naturally to a condition in which there is
no reason not to take according to your needs.
The essential, unbreakable connection between the give of production and the
take of consumption is broken naturally by the nature of software itself.
Fearing the economic consequences of this reality in a society which takes the
conservation of number and mass (natural properties of matter, but unnatural
for information), our capitalist societies have constructed elaborate,
centrally-administered market-controls (sound familiar?) in order to force the
information market, against its nature, to imitate the properties of the matter
marketplace.
We call these controls, collectively, the "intellectual property regime". And
what was once a fairly innocous implementation, limited in both time and scope,
has become enormous. When 12 year old kids and senior citizens are being
threatened with lawsuits and fines of more than they might earn in the next
ten years for the horrible crime of listening to music and sharing it with
their friends; when vast corporations use armies of lawyers to claim control of
trivial ideas through software patents; when international treaties hinge on
the application of stricter and stricter controls on the dissemination of
information; when the act of merely writing software capable of breaking
these manacles on inellectual freedom is made into a criminal offense
—can we really pretend that "IP laws" are any less oppressive than the
commuunist "command economy"?
Free software eliminates these unnatural controls, freeing the marketplace,
and allowing information products to assume their natural behavior. A
behavior, which —curiously— embraces the ideas of the communist
society.
In fact, the free/copyleft license strategy and the whole concept of
community-based peer production cus right across this political spectrum,
destroying the traditional boundaries, because it achieves the communist
ideals without the restrictions that capitalists object to.
As a social contract, free-copyleft licenses like the GPL draw a very
different boundary for personal "property" in the intellectual sphere than
either capitalism or communism does in the material world. Attribution is
accorded much greater importance, as the important fuel for the "reputation
game" that keeps excellent creators in a position to create. But the
ability to control the use and replication of the work is rejected. Indeed,
via the copyleft requirement, this artificial market control is denied to
all, ensuring that the work is free for use, reuse, improvement, and sharing
of the improved workd. The work is owned in that sense, not by any individual,
but by "the commons".
Thus it can be said to be both "communist" and "free-market" —at
least from a certain point of view. We in the West have been conditioned
to believe that the free-market always chooses competition, but in the case of
free software the free market chooses cooperation, or to put it more bluntly,
communism (that is to say, it encourages people to behave communally and hold
property in commin). People can participate in a free software project with
little or no "capital", so there is no reason to raise large capital
investments, hence no actual need for "capitallism" as such (as opposed to
the case of proprietary software, you do not need a company to start a free
software project!).
That, of course, is the cannonball that knocks the supports out from under the
capitalist arguments for the importance of intellectual property controls to
maintain production. They take it as a foregone conclusion that production
is impossible without monetary capital investment and that licensing-fees are
the only way to recoup that investment. However, free software production is
demonstrably so efficient and reductive to cost that capital investment
becomes almost irrelevant for all but a tiny minority of software projects.
Hence, less restrictive means of collecting revenue —such as service
contracts, commissions received beforehand, and investment based on personaluse value— have proved adequate to meet the much more modest funding
requirements of free software projects.
Contrary to the assumptions of our society, "free market" does not
necessarily imply "capitalist". "Capitalism" refers to the specific practice
of accumulating capital to start enterprises. In the material marketplace,
free markets appear to invariably lead to capitalism, but we can't safely
assume that will be so in information markets.
In the free-market of community-based peer production (CBPP), labor is
spontaneously contributed to the completion of goals for the pure joy (or
enlightened self-interest, if you prefer —which I generally do) of
helping the communal effort, based on the skill sets of the people
contributing. This happens completely without state coersion or market
controls. "From each according to his ability, to each according to his
need" occurs without any kind of state enforcement (i.e. in a free market).
Hence, CBPP or free software development, is a real, working case of "free
market communism".
A very intriguing approach, I think. Notice how Hancock limits this sort
of "free-market communism" (a concept, by the way, much in use back in the
1960s and 1970s, although forgotten ever since) to the software world, and
does not say a word about the "material market", where its practicality
remains doubtful. Still, it should be clear to anyone that free software
truly is inspired more by the values of classical Socialism than anything
else, although that does not necessarily imply that it is incompatible with
an open and free market at all. It just represents a given set of moral and
social values that, for the most part, are not fully compatible with the
world of selfish and ruthless competition that was supposed to become the norm
after the fall of the Berlin Wall.
{link to this story}
[Mon Aug 7 19:01:30 CEST 2006]
eWeek has a story on Bugle, a project that allows source code
queries using Google. The idea is simple, to be sure, but it certainly is
attractive too. Just check out their Google Source Code Bug Finder to get an idea.
{link to this story}