Everything in the list was something that many people
would think is good. They don't want advances, because
they have a hard enough time keeping up with what's
already out. They don't want foreigners anywhere
near them. They don't want any research into illnesses
they haven't personally suffered from.
To many people, it's all been a big success. To a select
few, 9/11 was the best thing that's happened in years.
Before it, they were worried about getting beat up over
a collapsing economy, corruption, and election fraud.
Now they're flying high, and all their friends are getting
enormous no-bid military procurement and reconstruction
contracts. Academic soreheads are easy to ignore.
Furthermore, I think NetBSD's (and OpenBSD's) UVM zero-copy
features are positively uttergloss. I wish Linux was poised
to offer anything even close. There's no way, though, that
Linux will have them before 2.8 or 3.0, in two to five years.
Making libstdc++ able to use those features transparently,
once they do appear, should hasten their arrival. There's
nothing like the prospect of making dozens (or hundreds) of
existing programs several times faster on millions of machines
to inspire kernel improvements. Having them already
demonstrably running faster on a competing OS helps too.
First, of course, we have to get those dozens (or hundreds)
of kernel-optimization-ready programs deployed, which means
making the improvements and getting a release that has them
out into the world. Fortunately the improvements are an
optimization even without UVM or its imagined Linux
equivalent, because they will speed up disk file I/O
operations too.
I hope they are not planning to stick with 3.3 for the
indefinite future. Gcc-3.4 is where the major improvements
are going, and its ABI is meant to be stable for a long time.
The 3.3 series is just for practice, as it were. For example,
getting iostreams to take advantage of NetBSD's UVM, and
expose zero-copy I/O at the user level, will happen early in
the 3.4 series. 3.4 is getting precompiled headers and other
practical work on faster compilation.
The same advice goes for Debian and the other distributions
as well (although of course Linux doesn't have UVM yet).
It would be a serious mistake to put in that much work just
for 3.3 itself, although the work isn't wasted because
after getting everything working on 3.3, switching to 3.4
should be (technically) pretty easy.
Efficiency of the whole system is more than just what
Carnot says. How much does a heliostat cost, per unit
of lifetime, vs. an area of greenhouse structure that
produces the same amount of power? The greenhouse is
bound to need a lot less maintenance. Never mind the
high-temperature parts of the heliostat tower, and the
pumps and pipes...
Google reports 200 MW for a 7 km diameter unit, to
go on line in 2005. They have plans to build four
more after that, no doubt adapting according to
what they learn from the first one. That's 38 km^2,
or 5 W/m^2. Doesn't sound like much, but how much
does a square meter of glass cost, out in the desert?
Australia has lots of desert.
Cost is quoted at A$800m (UKP 308m) for all five,
which is probably close to right. People have a
lot of experience with this kind of construction.
Is the cost per GW of capacity in line with other
renewable energy projects (dams, wind farms)? I don't know.
What, is that moon still there? Sometimes it seems like
things just go around and around the drain forever.
Seriously, it's funny how astronomers always think that
nobody will touch these rocks that are just sitting there
in handy orbits. It's the same with Cruithne, the asteroid
that co-orbits earth. They always say it will join Earth
again in 600 years (or whenever), and it never seems to cross
their minds that we might have found something more useful to
do with it by then.
Deimos will probably be more useful, though, than Phobos, as
a counterweight to attach to the end of the big elevator down
to the surface. We might have to move Phobos out of the way
-- making the elevator shimmy this way and that so that Phobos
just misses colliding each time past is asking for trouble.
Oddly enough, the convection tower, unlike the heliostat
tower described, is not a toy operated by an overfunded
federal research lab. Rather, the convection towers are
under construction for production use, having already been
proven on a smaller scale (also) in Spain. Therefore,
you may say they are not only on the radar screen of major
utilities, but are in fact on final approach for landing,
with passengers already lined up to board.
To describe the heliostat as "a bunch of swiveling mirrors"
would lead the reader to a drastic underestimation of the
difficulty of the project. The author of the paper touches
on many of the difficulties, including making
(one-of-a-kind) valves and pumps that will work on molten
salts. I didn't notice whether the likelihood of the
mirrors blowing away in a storm was touched on, or the
mirrors melting the top right off the tower if the salt
coolant ever stopped flowing. Their progress in
addressing these problems is commendable. What's not is
having spent the money to solve them unnecessarily when
better solutions are ready to hand.
In fact (and contrary to another poster's reply) the
convection towers continue to operate all night long.
Although they don't get insolation at night, the greenhouse
is still full of a huge amount of hot air, and the air
around it is suddenly cooler. Therefore, they provide their
own power storage, of sorts, another example of more-elegant
engineering. Of course the turbines may be started and
stopped at any time, as power needs dictate.
Finally, "retro" might refer to the 1950s engineering
approach embodied in the heliostat tower. The signature is
recognizable to anyone familiar with the history of public
technology projects. The space shuttle is another
example. Oil shale (anybody remember that?) was another.
Does anybody remember the story about the Crosspatch
Decision, from Heinlein's Stranger in a Strange Land
?
IBM has been very careful in the contracts it
signed, making careful distinction between ATT Unix
and IBM's own contributions. Not all Unix licensees were as
careful, possibly including Sequent. The question is, if Sequent's possibly incautious contract would have kept them
from contributing their inventions to Linux, would that
contract bind IBM, too? Would it prevent the inventions'
new owner from releasing them unencumbered? Or, do IBM's
own contracts with (the shell that is now) SCO subsume
Sequent's?
Yes, exercise, yes lists, yes meditation, yes get disconnected. If each alone doesn't help, you have to
combine them all and maybe something else besides.
I have found I also need to take my Ginkgo Biloba extract,
morning and night.
Ginkgo is described as promoting "ideation". That is a
way of saying that it helps when you suffer the peculiar
kind of depression not characterized by thoughts of suicide
or feelings of worthlessness, but rather, simply, that nothing
comes to mind. You sit and it just doesn't occur to you what
you might do. If you're in circumstances where you need to
look busy, you do passive things, like poking the reload
button on slashdot.
It takes a couple of weeks for Ginkgo to have any effect,
and it's always easiest to forget to take it when you need
it most. I keep it next to my toothbrush.
To file a bookmark, go down to the folder where
you want to file it, and right-click. There you go:
"Add Bookmark Here". "File Bookmark" was always
confusing. Not having it is an improvement.
In general, do a bit of exploration with the right
mouse button. You'll discover all kinds of good
things (besides a few remaining bugs).
It should say that Free Software projects would never
steal code from Microsoft. We Free Software authors take
pride in the reliability of our work, and Microsoft
code is known to crash frequently and to be susceptible
to viruses and internet worms. We don't want to adopt
those flaws into the Free Software world.
It's funny -- of the "top 3 bugs" each developer lists, I have
encountered only a couple. Of the top unimplemented features
they list, I don't want any of them. To me, the only really
critical bug was crashing within a few minutes every time I
turned on Javascript, but they might have licked that in the
14 July snapshot I run. After that, being able to
right-click on an image and get a menu reliably (not just
after the third try) would help. The only "new" feature
that would make much difference for me is honoring the Gnome
emacs key-binding preference. (ctrl-A, E, N, P, B, F, K,
particularly.) That worked fine in the 1.2 series, and is
the only feature I miss from it.
I have to admit that I'd like to see the per-site preference
behavior of cookies extended to Javascript, image loading and animation, font forcing, color forcing, zoom, etc. Probably
the most valuable improvement would be a way to use a
different text editor entirely, in another window if
necessary. But, mainly, it now has almost exactly the
feature set I need in a browser, and hardly anything else.
I wish it would stay that way.
I've been using 1.3 since the beginning, and it was pretty
sucky for a while, but I'm glad they did what they did. The
version I'm using now is so much nicer, all around, than
Epiphany, that it is clearly only politics that made Gnome
switch to the latter. I'll never switch, because the
Epiphany developers are a bunch of ideologues who have
announced they will never add the features that make the
browser useful and usable for me.
Wow, what a great telescope! If I could see that well,
I could spot a quark on Pluto from here without even
squinting.
Of course the article said 90 light years, which is
way too far to walk (or drive) anyway.
(We are 8 light minutes from the sun, and it would
take your whole life and part of your kids' to drive that far.)
I can't think of any reason why Microsoft would want
to sign a bootloader. That doesn't mean they won't. It just
means that persuasion may be needed. Consider that only one
bootloader (or anything that may be made to answer as a
bootloader) would be needed, and that they will have to
sign hundreds of things without being able to afford to
look very closely at them. Probably they have the legal
right to recall a product that turns out to contain a
bootloader, but once it's out, the necessary tracks will
be readily copyable to appropriate media along with a
whole bootable GNU or BSD distribution.
Probably if they turn out to have signed something they
regret, future versions of the console will recognize and
reject that signature (and owners of the affected CDs
would be obliged to trade them in, if they want to use
them on a newer console), but millions of existing consoles
would remain useful, and may achieve a second life after
MS has disowned them. It would probably be wise to keep
several signed loaders in reserve, so that when one is
repudiated, and the next version of the console released,
then the next loader may be pressed into service.
Those Martians are turning into a real menace! Not satisfied
with downing probes that approach their planet, they have
taken to shooting down craft as soon as they leave our own.
Join a Free Software project. Participate in bug fixing, at
first, and then cleanup, and then implementing new
features. After you develop some confidence and design
judgment, implement something substantial by yourself.
Then, do whatever it takes (meaning, really listen
to more experienced people) to get it accepted, and maintain
it and refine it according to qualified criticism and user
whims.
(Don't go start a new project. That's the last thing the
world needs, another abortive project run by a newbie.)
If programming turns out to be not your thing, you'll find
out soon enough, and before you've got yourself mired in.
One thing: as a Perl expert, you've most likely picked up
habits that would make you an awful programmer. You will
have to work hard to unlearn those.
When you go looking for professional programming work, you
can point to your substantial contributions, and they will
speak for you. Choose your project wisely.
Just etch a red pepper onto one or other of your buttocks
and count yourself adorned.
Anything more is a tool to help you filter out companies
you wouldn't want to be a part of, anyway, without wasting
too much time talking to them. Likewise potential mates,
potential mates' parents, bands, river-rafting clubs,
condo associations, military academies, supreme-judicial
internships, churches, university faculties, and diplomatic
appointments.
Alex Stepanov famously
described Java as "a money-oriented programming language".
I guess that makes the name "Jackpot" an appropriate name.
I suppose the next projects will be "Jingle" and "Jyp".
When you see a dancing bear, you don't evaluate how
well it dances, you marvel that it dances at all.
I suppose next we'll be asked if scooters are the
future of personal transportation, or talentless
teen-age models are the future of pop music, or invading countries that have oil is the future of world
commerce. (Oops!)
If you think the "no-fly" list is not working well,
you fundamentally misunderstand its purpose. In fact,
for its intended purpose, it has been working fabulously.
Like harrassing people about nail files in their bathroom
kits, it leads people to think that something is being
done about security, without the need actually to do
something. How many people have you heard say that while
having their shoes X-rayed was inconvenient, it made them
feel safer about flying?
X-raying shoes doesn't make for effective security, but it's
intrusive enough to give the impression that at least
something is being done.
Articles and editorials that call attention to the violations
that come with the bogus no-fly list are essential components
of the system -- they make everybody else experience it,
vicariously. Everybody who is a little bit stupid (i.e. most
people) feels a little safer for it. Sure it inconveniences
some people, but not enough to make much political difference.
Even better than the impression of intrusive security, it
leads to demands for what amounts to a system of internal
passports, where you can't travel by air without registering,
and getting -- and maintaining --- official permission.
"What, no internal passport? Sorry, sir, I can't let you
board." At first felons will have their passports pulled, then
"suspected terrorists", then political undesirables of all
sorts.
Whenever a corporate analyst issues a press release, it's
worth noting that corporate analysts hardly ever do anything
without having a customer P.R. department they can bill for
the time spent. The right question to ask is, who is the
plausible P.R. department customer for this Yankee analysis?
Former U.S. President Lyndon Johnson used to boast that he
had never cast an
unsold vote. He saw that as testament to his skill at finding
someone who wanted him to vote the "right way", and getting
a concession for it, which is how U.S. politics works.
It is possible for analysts to work that way too.
Aberdeen has been very good at finding customers for their
careful analyses, while Gartner appears to control expenses
by letting the customers do the writing. The analysts with a
shred of dignity left have recused themselves already because
they recognized that the NDA stacks the cards enough to
prevent any chance of anyone publishing a fair analysis.
The crash has been hard on analysts. (You can see that in
the periodic, contradictory swings by Gartner as IBM and
Microsoft alternately gain control of their corporate voice.)
Yankee, here, seems to be demonstrating mainly that they're
hungry.
Re:C++ and Kernel Coding
on
eCos 2.0 Released
·
· Score: 2, Interesting
Nowadays (what with Wince out there) we have to say
"hard real time". It delivers to-the-nanosecond
latency maxima, making it suitable for controlling
million-dollar machines that would break, and maybe
kill somebody, if it missed. It might be annoying
enough if it kept dropping your cellphone connection --
those have hard-real-time constraints, so when Wince
runs a cell phone, there's a separate CPU running a
real-time kernel.
There are hard-real-time kernels that will run underneath
Linux (or NetBSD) so you don't have to choose, you can
have both. Sometimes, though, you need networking but
can't afford the extra RAM and whatnot to run a whole
Unixy environment. Anyway the minimalism can be heady,
too, like the thin cold air on a mountain you've just
climbed.
Slashdot Mirror
To many people, it's all been a big success. To a select few, 9/11 was the best thing that's happened in years. Before it, they were worried about getting beat up over a collapsing economy, corruption, and election fraud. Now they're flying high, and all their friends are getting enormous no-bid military procurement and reconstruction contracts. Academic soreheads are easy to ignore.
Furthermore, I think NetBSD's (and OpenBSD's) UVM zero-copy features are positively uttergloss. I wish Linux was poised to offer anything even close. There's no way, though, that Linux will have them before 2.8 or 3.0, in two to five years.
Making libstdc++ able to use those features transparently, once they do appear, should hasten their arrival. There's nothing like the prospect of making dozens (or hundreds) of existing programs several times faster on millions of machines to inspire kernel improvements. Having them already demonstrably running faster on a competing OS helps too.
First, of course, we have to get those dozens (or hundreds) of kernel-optimization-ready programs deployed, which means making the improvements and getting a release that has them out into the world. Fortunately the improvements are an optimization even without UVM or its imagined Linux equivalent, because they will speed up disk file I/O operations too.
The same advice goes for Debian and the other distributions as well (although of course Linux doesn't have UVM yet). It would be a serious mistake to put in that much work just for 3.3 itself, although the work isn't wasted because after getting everything working on 3.3, switching to 3.4 should be (technically) pretty easy.
Google reports 200 MW for a 7 km diameter unit, to go on line in 2005. They have plans to build four more after that, no doubt adapting according to what they learn from the first one. That's 38 km^2, or 5 W/m^2. Doesn't sound like much, but how much does a square meter of glass cost, out in the desert? Australia has lots of desert.
Cost is quoted at A$800m (UKP 308m) for all five, which is probably close to right. People have a lot of experience with this kind of construction. Is the cost per GW of capacity in line with other renewable energy projects (dams, wind farms)? I don't know.
Seriously, it's funny how astronomers always think that nobody will touch these rocks that are just sitting there in handy orbits. It's the same with Cruithne, the asteroid that co-orbits earth. They always say it will join Earth again in 600 years (or whenever), and it never seems to cross their minds that we might have found something more useful to do with it by then.
Deimos will probably be more useful, though, than Phobos, as a counterweight to attach to the end of the big elevator down to the surface. We might have to move Phobos out of the way -- making the elevator shimmy this way and that so that Phobos just misses colliding each time past is asking for trouble.
To describe the heliostat as "a bunch of swiveling mirrors" would lead the reader to a drastic underestimation of the difficulty of the project. The author of the paper touches on many of the difficulties, including making (one-of-a-kind) valves and pumps that will work on molten salts. I didn't notice whether the likelihood of the mirrors blowing away in a storm was touched on, or the mirrors melting the top right off the tower if the salt coolant ever stopped flowing. Their progress in addressing these problems is commendable. What's not is having spent the money to solve them unnecessarily when better solutions are ready to hand.
In fact (and contrary to another poster's reply) the convection towers continue to operate all night long. Although they don't get insolation at night, the greenhouse is still full of a huge amount of hot air, and the air around it is suddenly cooler. Therefore, they provide their own power storage, of sorts, another example of more-elegant engineering. Of course the turbines may be started and stopped at any time, as power needs dictate.
Finally, "retro" might refer to the 1950s engineering approach embodied in the heliostat tower. The signature is recognizable to anyone familiar with the history of public technology projects. The space shuttle is another example. Oil shale (anybody remember that?) was another.
IBM has been very careful in the contracts it signed, making careful distinction between ATT Unix and IBM's own contributions. Not all Unix licensees were as careful, possibly including Sequent. The question is, if Sequent's possibly incautious contract would have kept them from contributing their inventions to Linux, would that contract bind IBM, too? Would it prevent the inventions' new owner from releasing them unencumbered? Or, do IBM's own contracts with (the shell that is now) SCO subsume Sequent's?
I have found I also need to take my Ginkgo Biloba extract, morning and night.
Ginkgo is described as promoting "ideation". That is a way of saying that it helps when you suffer the peculiar kind of depression not characterized by thoughts of suicide or feelings of worthlessness, but rather, simply, that nothing comes to mind. You sit and it just doesn't occur to you what you might do. If you're in circumstances where you need to look busy, you do passive things, like poking the reload button on slashdot.
It takes a couple of weeks for Ginkgo to have any effect, and it's always easiest to forget to take it when you need it most. I keep it next to my toothbrush.
In general, do a bit of exploration with the right mouse button. You'll discover all kinds of good things (besides a few remaining bugs).
It should say that Free Software projects would never steal code from Microsoft. We Free Software authors take pride in the reliability of our work, and Microsoft code is known to crash frequently and to be susceptible to viruses and internet worms. We don't want to adopt those flaws into the Free Software world.
I have to admit that I'd like to see the per-site preference behavior of cookies extended to Javascript, image loading and animation, font forcing, color forcing, zoom, etc. Probably the most valuable improvement would be a way to use a different text editor entirely, in another window if necessary. But, mainly, it now has almost exactly the feature set I need in a browser, and hardly anything else. I wish it would stay that way.
I've been using 1.3 since the beginning, and it was pretty sucky for a while, but I'm glad they did what they did. The version I'm using now is so much nicer, all around, than Epiphany, that it is clearly only politics that made Gnome switch to the latter. I'll never switch, because the Epiphany developers are a bunch of ideologues who have announced they will never add the features that make the browser useful and usable for me.
Of course the article said 90 light years, which is way too far to walk (or drive) anyway. (We are 8 light minutes from the sun, and it would take your whole life and part of your kids' to drive that far.)
Probably if they turn out to have signed something they regret, future versions of the console will recognize and reject that signature (and owners of the affected CDs would be obliged to trade them in, if they want to use them on a newer console), but millions of existing consoles would remain useful, and may achieve a second life after MS has disowned them. It would probably be wise to keep several signed loaders in reserve, so that when one is repudiated, and the next version of the console released, then the next loader may be pressed into service.
(By the way, others reading the parent posting read that line with its intended humor.)
We must declare war on adventurist protectionism.
Mars delenda est!
There's only one thing I'd like to see changed between RC3 and release: make it stop crashing within minutes any time I turn on Javascript.
(Don't go start a new project. That's the last thing the world needs, another abortive project run by a newbie.)
If programming turns out to be not your thing, you'll find out soon enough, and before you've got yourself mired in. One thing: as a Perl expert, you've most likely picked up habits that would make you an awful programmer. You will have to work hard to unlearn those.
When you go looking for professional programming work, you can point to your substantial contributions, and they will speak for you. Choose your project wisely.
Anything more is a tool to help you filter out companies you wouldn't want to be a part of, anyway, without wasting too much time talking to them. Likewise potential mates, potential mates' parents, bands, river-rafting clubs, condo associations, military academies, supreme-judicial internships, churches, university faculties, and diplomatic appointments.
Alex Stepanov famously described Java as "a money-oriented programming language". I guess that makes the name "Jackpot" an appropriate name. I suppose the next projects will be "Jingle" and "Jyp".
I suppose next we'll be asked if scooters are the future of personal transportation, or talentless teen-age models are the future of pop music, or invading countries that have oil is the future of world commerce. (Oops!)
X-raying shoes doesn't make for effective security, but it's intrusive enough to give the impression that at least something is being done.
Articles and editorials that call attention to the violations that come with the bogus no-fly list are essential components of the system -- they make everybody else experience it, vicariously. Everybody who is a little bit stupid (i.e. most people) feels a little safer for it. Sure it inconveniences some people, but not enough to make much political difference.
Even better than the impression of intrusive security, it leads to demands for what amounts to a system of internal passports, where you can't travel by air without registering, and getting -- and maintaining --- official permission. "What, no internal passport? Sorry, sir, I can't let you board." At first felons will have their passports pulled, then "suspected terrorists", then political undesirables of all sorts.
Former U.S. President Lyndon Johnson used to boast that he had never cast an unsold vote. He saw that as testament to his skill at finding someone who wanted him to vote the "right way", and getting a concession for it, which is how U.S. politics works. It is possible for analysts to work that way too. Aberdeen has been very good at finding customers for their careful analyses, while Gartner appears to control expenses by letting the customers do the writing. The analysts with a shred of dignity left have recused themselves already because they recognized that the NDA stacks the cards enough to prevent any chance of anyone publishing a fair analysis.
The crash has been hard on analysts. (You can see that in the periodic, contradictory swings by Gartner as IBM and Microsoft alternately gain control of their corporate voice.) Yankee, here, seems to be demonstrating mainly that they're hungry.
Lambda library, compile-time parser generator library, graph algorithms, regular expressions, safe threads, and on and on.
There are hard-real-time kernels that will run underneath Linux (or NetBSD) so you don't have to choose, you can have both. Sometimes, though, you need networking but can't afford the extra RAM and whatnot to run a whole Unixy environment. Anyway the minimalism can be heady, too, like the thin cold air on a mountain you've just climbed.
eCos serves as elegant proof that, even in the Free Software world, C++ is a practical language to use for even the lowest-level kernel coding.