The more I read by and about Ken Brown, the more he resembles Mike Vandeman, the legendary kook who used to harass mountain bikers in rec.bicycles.off-road. He used the same procedure:
start with a conclusion
collect facts
ignore the facts that contradict the conclusion
when no supporting facts remain, affirm the conclusion
"On the other hand, IMO your budget is way low....
"My recommendation? Up your budget quite a bit. Check out the Canon Digital Rebel. Yes, its about $1k with a pretty good generic lens. But that may be less than you'd spend over a year with a $200-300 film camera, plus decent film, plus developing. Think TCO not just initial purchase price."
That's just about the worst advice you can give a new photographer. For one thing, you can buy an awful lot of film and processing for that $700-$800 difference. For another, it's better to enter a hobby at the bottom in case you discover that don't like it or don't have as much time for it as you expected to. That way, your money isn't tied up in an expensive paperweight.
Now, if you had suggested buying an inexpensive digital camera, maybe a point-and-shoot, then I wouldn't disagree with you as strongly. I'd see that you can use it to learn about composition:
while (true) { take a picture examine composition in the LCD decide how to improve composition }
Even then, I might still disagree with you if I knew that the new photographer wanted to make prints at 8x10 or larger. Until you get into the professional equipment stratosphere (i.e. medium- and large-format digital, thousands of dollars) film still rules in that respect.
Never buy a photographer a camera as a gift unless you know exactly what the photographer wants or it's the photographer's first camera and you are her mentor or have the advice of her mentor.
Photographers are fussy about cameras, lenses, and accessories.
The safest thing to buy for a photographer is film (or memory cards for those who shoot digital).
Sweet. One-speed bikes are bikes distilled to the essential components. They're so elegant.
I made my one-speed out of 1986 Univega Vivasport. It was my first "nice" bike, the first I'd ever owned that weighed less than 30 pounds. I wasn't ready to commit to fixed-gear, so it has a freewheel. I'm pretty happy about it.
>And 40 pounds for a complete bike of this type is fairly light to be honest.
That's 40 pounds for a gravity bike, a bike you can only ride downhill. To get it to the top of the mountain, you have to stick it on a chair lift. Where's the fun in that?
OK, I installed xfce4. xfdesktop manages the root menu. That's good. It looks like xfwm4 manages keystrokes. That's not good. It's not as bad as, say, watching your cat get hit by a car, but it's not the way I want my window manager to behave.
The last time I looked at xfce, there was no separate program for menu and for translating keystrokes into window manager events. I'll give xfce 4 a closer look.
I agree about that but I think that all modern window managers get it wrong. They aren't built according to the UNIX philosophy of small programs that work together.
Window managers should manage (decorate, place, tile, open, close, resize) windows. They shouldn't manage desktops, deal with icons, handle keyboard mappings, or provide an application menu. Those actions are the jobs of other programs.
If there was a window manager and a suite of supporting programs, then you could have more choices about the way you work. You could choose the best of breed, the supporting programs that work the way that you want.
"computron" has been used since at least the mid-1980s, when I first heard it used by an MIT graduate.
From Jargon File (4.0.0/24 July 1996) [jargon]:
computron/kom'pyoo-tron`//n./ 1. A notional unit of
computing power combining instruction speed and storage capacity,
dimensioned roughly in instructions-per-second times
megabytes-of-main-store times megabytes-of-mass-storage. "That
machine can't run GNU Emacs, it doesn't have enough computrons!"
This usage is usually found in metaphors that treat computing power
as a fungible commodity good, like a crop yield or diesel
horsepower. See {bitty box}, {Get a real computer!},
{toy}, {crank}. 2. A mythical subatomic particle that bears
the unit quantity of computation or information, in much the same
way that an electron bears one unit of electric charge (see also
{bogon}). An elaborate pseudo-scientific theory of computrons
has been developed based on the physical fact that the molecules in
a solid object move more rapidly as it is heated. It is argued
that an object melts because the molecules have lost their
information about where they are supposed to be (that is, they have
emitted computrons). This explains why computers get so hot and
require air conditioning; they use up computrons. Conversely, it
should be possible to cool down an object by placing it in the path
of a computron beam. It is believed that this may also explain why
machines that work at the factory fail in the computer room: the
computrons there have been all used up by the other hardware.
(This theory probably owes something to the "Warlock" stories
by Larry Niven, the best known being "What Good is a Glass
Dagger?", in which magic is fueled by an exhaustible natural
resource called `mana'.)
It never ceases to amaze me that people continue to try to sell motors for bicycles. I can see the appeal of a moped or a motorcycle, and while they have their own merits, they are completely different vehicles from the bicycle.
The beauty of the bike is that it's efficient, it's clean, it's quiet, and regular riding improves or maintains the health of the rider. It can be cheap to own and maintain.
Maybe a few people need motors for independence. I can see the elderly and some handicapped people using them. For the rest, I think this sums the potential of the motorized bicycle:
repeat { Gas up the 'bike' Ride for an hour } until (the novelty wears off)
Back in the day, a Nevada company called Stride made a microcomputer with a head-mounted input device called "the Nod". ISTR that the computer was a 68K system and that Jerry Pournell of _Byte_ was enamored by the thing.
Bicyclists have known about Ti for years
on
The Sexiest Metal
·
· Score: 1
Of course it's sexy. It gets you your Glow In the Dark points, too.:-)
Back in the day when people had real choices to make over the personal computers they bought, the conventional wisdom said, "Buy the computer that runs the application(s) you must run."
Back in the day before that, when there were no personal computers, computers took filled closets or rooms, and programmers were really mathematicians and physicists and linguists, hardware companies gave away software, in part so you'd buy their hardware.
Sun's decision to make StarOffice "free" for licensees of Solaris makes sense in this light.
These days, we call it "Value Added".
I want books with plenty of examples. Start with simple ones. Work up to complex ones that I can use in the real world.
The O'Reilly Bison book is an example of a technical book without sufficient examples. It's little more than a bound version of the original Bison documentation. The toy examples, a desk calculator, for instance, don't come close to the kinds of tricks you need to process real languages. Useful examples would be things like using flex and bison with C++, which alas isn't really documented anywhere, using bison to parse strings (i.e. in memory, not in a file), uses for semantic values, manipulating yyinput to parse more than one file, parsing non-LR languages. The last one is tricky. I once wrote a COBOL compiler using flex and bison. It's possible!
One of my co-workers just taught himself to use autoconf and automake. He told me that the examples in those books are simple and contrived. They don't touch what you really have to do to fit your package into autoconf and automake.
Back in the day, I joined a small startup. Their policy was to make the new guy, the most junior one, the sysadmin. It was SunOS on a network of Sun 3/50s and 3/60s with a 3/1xx file server. Our connection to the outside world was UUCP over a modem, so we didn't have too much to worry about in terms of external security.
Before that, I worked full-time in academic computing, both as a programmer and as a junior sysadmin for a VAX/VMS network with occasional care and feeding of an AT&T 3B2. Mostly, I set up user accounts, installed software, and made cables for the terminals we used.
I had (still have) a CS degree. I found that it was hard to escape the sysadmin ghetto for a job as a programmer. Leaving the ghetto was the best thing I ever did for my career. It was fun at times. Sometimes I was the hero but most of the times I was the goat. People don't thank their sysadmins for making the computers and the network run smoothly. People always bitch at their sysadmins when something goes wrong.
I gave up ganja in college. One night I was smoking with some friends and I wondered who was peeking over my shoulder. I looked and discovered that I was sitting with my back against the wall. I stopped right there.
I experimented with minor hallucenogens after that but they took hours to wear off.
Right now, I make my living being smart. Anything more recreational than an occasional beer or daily cups of coffee require that I take my brain off line for too long. I don't like feeling stupid.
All that cool stuff that NASA proposes is 20 years away. That's what NASA does. It captures the American fascination with gadgets and then gets Congress to fund our collective technology jones.
If you want to see what can be done in 2 years instead of 20, take a look at what we're doing at
TGV Rockets. When we're fully funded, our reusable, suborbital rockets will open the door to cheap access to space. I'm talking about $1000 per kilogram instead of the $20000 per kilogram you pay today.
G. Harry Stine wrote about Single Stage to Orbit (SSTO) rockets taking us half way to anywere. Making reusable, suborbital rockets first is the baby step we absolutely must take before we can achieve SSTO and then the stars.
Slashdot Mirror
The more I read by and about Ken Brown, the more he resembles Mike Vandeman, the legendary kook who used to harass mountain bikers in rec.bicycles.off-road. He used the same procedure:
- start with a conclusion
- collect facts
- ignore the facts that contradict the conclusion
- when no supporting facts remain, affirm the conclusion
There is a FAQ about MV.>War is terrible. Games are fun. Ne'er the two should meet. IMHO.
Like in
Ender's Game ?
"On the other hand, IMO your budget is way low. ...
"My recommendation? Up your budget quite a bit. Check out the Canon Digital Rebel. Yes, its about $1k with a pretty good generic lens. But that may be less than you'd spend over a year with a $200-300 film camera, plus decent film, plus developing. Think TCO not just initial purchase price."
That's just about the worst advice you can give a new photographer. For one thing, you can buy an awful lot of film and processing for that $700-$800 difference. For another, it's better to enter a hobby at the bottom in case you discover that don't like it or don't have as much time for it as you expected to. That way, your money isn't tied up in an expensive paperweight.
Now, if you had suggested buying an inexpensive digital camera, maybe a point-and-shoot, then I wouldn't disagree with you as strongly. I'd see that you can use it to learn about composition:
Even then, I might still disagree with you if I knew that the new photographer wanted to make prints at 8x10 or larger. Until you get into the professional equipment stratosphere (i.e. medium- and large-format digital, thousands of dollars) film still rules in that respect.
Never buy a photographer a camera as a gift unless you know exactly what the photographer wants or it's the photographer's first camera and you are her mentor or have the advice of her mentor.
Photographers are fussy about cameras, lenses, and accessories.
The safest thing to buy for a photographer is film (or memory cards for those who shoot digital).
Sweet. One-speed bikes are bikes distilled to the essential components. They're so elegant.
I made my one-speed out of 1986 Univega Vivasport. It was my first "nice" bike, the first I'd ever owned that weighed less than 30 pounds. I wasn't ready to commit to fixed-gear, so it has a freewheel. I'm pretty happy about it.
That's 40 pounds for a gravity bike, a bike you can only ride downhill. To get it to the top of the mountain, you have to stick it on a chair lift. Where's the fun in that?
Sounds like a fine, fun idea. We can use it to educate the masses and reclaim the word. I stuck the logo on my own web site
OK, I installed xfce4. xfdesktop manages the root menu. That's good. It looks like xfwm4 manages keystrokes. That's not good. It's not as bad as, say, watching your cat get hit by a car, but it's not the way I want my window manager to behave.
The last time I looked at xfce, there was no separate program for menu and for translating keystrokes into window manager events. I'll give xfce 4 a closer look.
I agree about that but I think that all modern window managers get it wrong. They aren't built according to the UNIX philosophy of small programs that work together.
Window managers should manage (decorate, place, tile, open, close, resize) windows. They shouldn't manage desktops, deal with icons, handle keyboard mappings, or provide an application menu. Those actions are the jobs of other programs.
If there was a window manager and a suite of supporting programs, then you could have more choices about the way you work. You could choose the best of breed, the supporting programs that work the way that you want.
That puts things in perspective for me. When my friends think I'm odd for spending $2000 on a bicycle, I'll tell them about the $400K 959.
"computron" has been used since at least the mid-1980s, when I first heard it used by an MIT graduate.
/kom'pyoo-tron`/ /n./ 1. A notional unit of
From Jargon File (4.0.0/24 July 1996) [jargon]:
computron
computing power combining instruction speed and storage capacity,
dimensioned roughly in instructions-per-second times
megabytes-of-main-store times megabytes-of-mass-storage. "That
machine can't run GNU Emacs, it doesn't have enough computrons!"
This usage is usually found in metaphors that treat computing power
as a fungible commodity good, like a crop yield or diesel
horsepower. See {bitty box}, {Get a real computer!},
{toy}, {crank}. 2. A mythical subatomic particle that bears
the unit quantity of computation or information, in much the same
way that an electron bears one unit of electric charge (see also
{bogon}). An elaborate pseudo-scientific theory of computrons
has been developed based on the physical fact that the molecules in
a solid object move more rapidly as it is heated. It is argued
that an object melts because the molecules have lost their
information about where they are supposed to be (that is, they have
emitted computrons). This explains why computers get so hot and
require air conditioning; they use up computrons. Conversely, it
should be possible to cool down an object by placing it in the path
of a computron beam. It is believed that this may also explain why
machines that work at the factory fail in the computer room: the
computrons there have been all used up by the other hardware.
(This theory probably owes something to the "Warlock" stories
by Larry Niven, the best known being "What Good is a Glass
Dagger?", in which magic is fueled by an exhaustible natural
resource called `mana'.)
Hey, DisKurzion, go visit the web site. There's a choice for DirectTV.
This is really a question for Darick Robertson but he's not here.
Why are Yelena and Channon drawn differently in the final issue? They're lean in all of the previous issues and kinda puffy in the final issue.
It never ceases to amaze me that people continue to try to sell motors for bicycles. I can see the appeal of a moped or a motorcycle, and while they have their own merits, they are completely different vehicles from the bicycle.
The beauty of the bike is that it's efficient, it's clean, it's quiet, and regular riding improves or maintains the health of the rider. It can be cheap to own and maintain.
Maybe a few people need motors for independence. I can see the elderly and some handicapped people using them. For the rest, I think this sums the potential of the motorized bicycle:
An equally interesting question is: Are the SunRays compatible with vnc?
Back in the day, a Nevada company called Stride made a microcomputer with a head-mounted input device called "the Nod". ISTR that the computer was a 68K system and that Jerry Pournell of _Byte_ was enamored by the thing.
Of course it's sexy. It gets you your Glow In the Dark points, too. :-)
For examples of Ti bikes, see Litespeed, Merlin, and Santana Tandems.
Back in the day when people had real choices to make over the personal computers they bought, the conventional wisdom said, "Buy the computer that runs the application(s) you must run."
Back in the day before that, when there were no personal computers, computers took filled closets or rooms, and programmers were really mathematicians and physicists and linguists, hardware companies gave away software, in part so you'd buy their hardware.
Sun's decision to make StarOffice "free" for licensees of Solaris makes sense in this light.
These days, we call it "Value Added".
I want books with plenty of examples. Start with simple ones. Work up to complex ones that I can use in the real world.
The O'Reilly Bison book is an example of a technical book without sufficient examples. It's little more than a bound version of the original Bison documentation. The toy examples, a desk calculator, for instance, don't come close to the kinds of tricks you need to process real languages. Useful examples would be things like using flex and bison with C++, which alas isn't really documented anywhere, using bison to parse strings (i.e. in memory, not in a file), uses for semantic values, manipulating yyinput to parse more than one file, parsing non-LR languages. The last one is tricky. I once wrote a COBOL compiler using flex and bison. It's possible!
One of my co-workers just taught himself to use autoconf and automake. He told me that the examples in those books are simple and contrived. They don't touch what you really have to do to fit your package into autoconf and automake.
Back in the day, I joined a small startup. Their policy was to make the new guy, the most junior one, the sysadmin. It was SunOS on a network of Sun 3/50s and 3/60s with a 3/1xx file server. Our connection to the outside world was UUCP over a modem, so we didn't have too much to worry about in terms of external security.
Before that, I worked full-time in academic computing, both as a programmer and as a junior sysadmin for a VAX/VMS network with occasional care and feeding of an AT&T 3B2. Mostly, I set up user accounts, installed software, and made cables for the terminals we used.
I had (still have) a CS degree. I found that it was hard to escape the sysadmin ghetto for a job as a programmer. Leaving the ghetto was the best thing I ever did for my career. It was fun at times. Sometimes I was the hero but most of the times I was the goat. People don't thank their sysadmins for making the computers and the network run smoothly. People always bitch at their sysadmins when something goes wrong.
Why bother securing the network? Secure the hosts and then you can use them safely on any network, trusted or untrusted.
I gave up ganja in college. One night I was smoking with some friends and I wondered who was peeking over my shoulder. I looked and discovered that I was sitting with my back against the wall. I stopped right there.
I experimented with minor hallucenogens after that but they took hours to wear off.
Right now, I make my living being smart. Anything more recreational than an occasional beer or daily cups of coffee require that I take my brain off line for too long. I don't like feeling stupid.
All that cool stuff that NASA proposes is 20 years away. That's what NASA does. It captures the American fascination with gadgets and then gets Congress to fund our collective technology jones.
If you want to see what can be done in 2 years instead of 20, take a look at what we're doing at TGV Rockets. When we're fully funded, our reusable, suborbital rockets will open the door to cheap access to space. I'm talking about $1000 per kilogram instead of the $20000 per kilogram you pay today.
G. Harry Stine wrote about Single Stage to Orbit (SSTO) rockets taking us half way to anywere. Making reusable, suborbital rockets first is the baby step we absolutely must take before we can achieve SSTO and then the stars.