Does DSL's speed guarantee come historically directly from ATM? I seem to recall that ATM had QoS guarantees, and I was wondering whether the speed provisions of DSL (you always get your rated speed) were designed in on purpose, or were already there as an unintended bonus of the pre-existing phone networks?
Isn't the point of a single system to have exactly identical machines? You either mount / over NFS or have an image you blast onto the machines regularly. You absolutely mount/home over NFS, and use local disk only for transient, unbacked up data --- like builds, scratch, swap.
I think HP was right on the money; allowing your infrastructure to heterogenize is a bad thing.
Note that you can still have dept-local configurations, as long as they are centrally managed images. The idea is that if any machine breaks, you can take it away and replace it without any loss (modulo the transient files) to the user.
Of course, I've never managed more than 3 machines, so I'm probably missing something: what?
Can anyone explain where the buildings for City come from? I get most of the description, but the buildings seem to not be enumerated or randomly generated anywhere.
I shelled out the extra cash for the IR enabled nokia 6100, back in the day (also cingular -- or whatever they were called before that).
I never did use the IR, which is just as good, as when I finally disassembled the phone (after dunking it in a cooler) I discovered that there was no IR tx, just the darkened peice of plastic where it should go...
I can never predict how thunderbird is going to mangle my text formatting.
I use a monospace font for a reason, so that I can make decent bulleted lists (I would kill for filladapt) or cheap ascii-art. it seems that half the time the text editor decides to override me and manually wrap before sending... which looks kinda bad.
I get no value from sending html, as I try not to alienate eveyone I email.
... in that case, outlook must be CRAP. I am a long time emacs-MH user, and I find thunderbird (which I switched to in order to interact better with the increasing tech-unsavvvy family) to be slow, unreliable (over IMAP at least, it requires several restarts daily to get its state synchronized with the vanilla imap server we use u-wisc?), offer bad text formatting, and spotty rendering of some emails.
in short, Thunderbird SUCKS.
KMail will be the next test victim, after that.. I may have to revert to emacs-mh. I mean 99.9% of email is text, so why do email client have such crap text editing engines?
there are so many patents out there that have patented the "desire" not the "implementation".
* I'd like to make shopping on the web easy enough to require only one click!
* I'd like to compress my images on a camera!
* I'd like to have richly marked up pages stored on a server and allow remote clients to access and display them.
These are NOT patent worthy, any more than
* I'd like a non-fire-based, perpetual source of light.
However, the invention of
* carbon filament in a noble gas through which electricity is passed to cause incandesence
IS a patentable invention, because it is a novel IMPLEMENTATION, not a novel desire.... or rather, above I should have said "... SHOULD not be patent worthy." I'm sure the system has progressed far beyond its original intentions.
While I'm on the subject, I'd like to question whether any patents today actually function to foster innovation. It seems that the strategic patent is a very potent weapon AGAINST innovation (on the part of your competitors) by locking others out of a business endeavor you have no intention to persue--yet--but want to keep others from. See the drug industry for rampant examples: They patent the drug, and just when that's about to expire, they patent the messenger whatsit, and when that expires, they patent the resulting whosit... some of which are produced naturally by your body. Hell, they've even started patenting DNA, which they don't even know what it DOES!
At some point, you have to stop swallowing the propaganda and question these things.
I worked with VSS and Visual Age. A check in went like this:
1) lock entire VSS source tree 2) export VSS onto your HD. 3) export your Visual Age source ontop of the VSS source tree 4) use another versioning tool to tell you which files you've touched. 5) check only these files back into VSS 6) unlock.
A complete PITA, basically because we were using two repositories in parallel. Every so often, they would fall out of sync, or someone would get the order wrong and go backwards in time.
perhaps build a root ramdisk with the most commonly used apps? Then you could spin down and unmount your hd partitions. 'course you would lose your data if you unexpectedly kill the laptop.
It may be the winmodem, or some other win-peripheral (sp?). IIRC, these load the batter a bit, and if they default to on, that would be a problem.
But you probably don't need always-on: indeed, the single orbiter may be sufficient when combined with inertial systems onboard the rover.
The reason one satelite should be enough is that I'm guesing you can just take distance measurements over a period of time. Thus, the one satelite can function as three, as long as you are willing to wait for the reading and can guestimate how you've travelled in the meantime.
MY GF has a cheapo machine her brother in law gave her. In order to use a browser (either EI or firebird) she must open a cmd window and ping some internet host.
failure to do this before opening the browser has the machine die on a service related error "the rpc service has unexpectedly shut down. rebooting".
I didn't see any references to the efficiency of the turbine, but I doubt he's claiming they are 100% efficient. In fact, in one paragraph, he points out that if the turbines didn't manage to regain enough energy to compress the gas, the plane could be landed and pointed into the wind, and thus recharge itself.
The wind is thus input, and it isn't perpetual motion.
... how do they undo state mutation? Do they keep an undo/redo stack for the heap? eww...
I always liked smalltalk's code browser that let you modify your methods while the program was running; this lead to some really fast development.
I'm wondering whether your experience w/ the hot-swap enabled debugger is the same: that most development is done while debugging. You just stub out your methods, start the program, and add code until it's running on your sample input. (then you sit down and refactor) Or does the system-programming nature of java come through, that you sit down and agonize for a week over which classes do what and where the methods go and what your invariants and edge cases are?
Man, I enjoyed smalltalk (digitalk smalltalk for the mac: ran in less than half a meg, including a GUI widget set)
Slashdot Mirror
Does DSL's speed guarantee come historically directly from ATM? I seem to recall that ATM had QoS guarantees, and I was wondering whether the speed provisions of DSL (you always get your rated speed) were designed in on purpose, or were already there as an unintended bonus of the pre-existing phone networks?
confusion.
/home over NFS, and use local disk only for transient, unbacked up data --- like builds, scratch, swap.
Isn't the point of a single system to have exactly identical machines? You either mount / over NFS or have an image you blast onto the machines regularly. You absolutely mount
I think HP was right on the money; allowing your infrastructure to heterogenize is a bad thing.
Note that you can still have dept-local configurations, as long as they are centrally managed images. The idea is that if any machine breaks, you can take it away and replace it without any loss (modulo the transient files) to the user.
Of course, I've never managed more than 3 machines, so I'm probably missing something: what?
thanks. that's the hint I needed.
*shudder*
I can't imagine why anyone would willingly code in C++. It seems at times to be specifically designed to write hard-to-predict code.
Not that Java is much better, but at least this sort of ill-behavior is hard to acheive.
So make round pillars with the fibers laid out so that they always go perpendicular to the surface tangent.
You could put a steel pillar in the middle -- as long as the fibers go around it.
Can anyone explain where the buildings for City come from? I get most of the description, but the buildings seem to not be enumerated or randomly generated anywhere.
I shelled out the extra cash for the IR enabled nokia 6100, back in the day (also cingular -- or whatever they were called before that).
I never did use the IR, which is just as good, as when I finally disassembled the phone (after dunking it in a cooler) I discovered that there was no IR tx, just the darkened peice of plastic where it should go...
what a gyp!
I can never predict how thunderbird is going to mangle my text formatting.
I use a monospace font for a reason, so that I can make decent bulleted lists (I would kill for filladapt) or cheap ascii-art. it seems that half the time the text editor decides to override me and manually wrap before sending... which looks kinda bad.
I get no value from sending html, as I try not to alienate eveyone I email.
pinky?
try algernon.
... in that case, outlook must be CRAP. I am a long time emacs-MH user, and I find thunderbird (which I switched to in order to interact better with the increasing tech-unsavvvy family) to be slow, unreliable (over IMAP at least, it requires several restarts daily to get its state synchronized with the vanilla imap server we use u-wisc?), offer bad text formatting, and spotty rendering of some emails.
in short, Thunderbird SUCKS.
KMail will be the next test victim, after that.. I may have to revert to emacs-mh. I mean 99.9% of email is text, so why do email client have such crap text editing engines?
linksys?
no!
... or rather, above I should have said "... SHOULD not be patent worthy." I'm sure the system has progressed far beyond its original intentions.
there are so many patents out there that have patented the "desire" not the "implementation".
* I'd like to make shopping on the web easy enough to require only one click!
* I'd like to compress my images on a camera!
* I'd like to have richly marked up pages stored on a server and allow remote clients to access and display them.
These are NOT patent worthy, any more than
* I'd like a non-fire-based, perpetual source of light.
However, the invention of
* carbon filament in a noble gas through which electricity is passed to cause incandesence
IS a patentable invention, because it is a novel IMPLEMENTATION, not a novel desire.
While I'm on the subject, I'd like to question whether any patents today actually function to foster innovation. It seems that the strategic patent is a very potent weapon AGAINST innovation (on the part of your competitors) by locking others out of a business endeavor you have no intention to persue--yet--but want to keep others from. See the drug industry for rampant examples: They patent the drug, and just when that's about to expire, they patent the messenger whatsit, and when that expires, they patent the resulting whosit... some of which are produced naturally by your body. Hell, they've even started patenting DNA, which they don't even know what it DOES!
At some point, you have to stop swallowing the propaganda and question these things.
yup.
I worked with VSS and Visual Age. A check in went like this:
1) lock entire VSS source tree
2) export VSS onto your HD.
3) export your Visual Age source ontop of the VSS source tree
4) use another versioning tool to tell you which files you've touched.
5) check only these files back into VSS
6) unlock.
A complete PITA, basically because we were using two repositories in parallel. Every so often, they would fall out of sync, or someone would get the order wrong and go backwards in time.
perhaps build a root ramdisk with the most commonly used apps? Then you could spin down and unmount your hd partitions. 'course you would lose your data if you unexpectedly kill the laptop.
It may be the winmodem, or some other win-peripheral (sp?). IIRC, these load the batter a bit, and if they default to on, that would be a problem.
But you probably don't need always-on: indeed, the single orbiter may be sufficient when combined with inertial systems onboard the rover.
The reason one satelite should be enough is that I'm guesing you can just take distance measurements over a period of time. Thus, the one satelite can function as three, as long as you are willing to wait for the reading and can guestimate how you've travelled in the meantime.
I meant ping ANY internet computer.
Thanks for the suggestion, tho. Her AV claims the computer is clean.
set theory? nah; plumbing!
monads are plumbing.
OT, but maybe one of you has a suggestion
MY GF has a cheapo machine her brother in law gave her. In order to use a browser (either EI or firebird) she must open a cmd window and ping some internet host.
failure to do this before opening the browser has the machine die on a service related error "the rpc service has unexpectedly shut down. rebooting".
Any ideas on why and how to stop this?
sooo...
it seems he's put the servos inside the rotor head, and has replaced the swashplate with electrical connections.
Seems kinda weird tho. can servos really react that quickly?
It's just a phone cable, so perhaps a dummy jack in the back and a cable to the internal card?
Next we break out the bailing wire and chewing gum to hold the case together.
I didn't see any references to the efficiency of the turbine, but I doubt he's claiming they are 100% efficient. In fact, in one paragraph, he points out that if the turbines didn't manage to regain enough energy to compress the gas, the plane could be landed and pointed into the wind, and thus recharge itself.
The wind is thus input, and it isn't perpetual motion.
full height + 90 degree riser?
... how do they undo state mutation? Do they keep an undo/redo stack for the heap? eww...
I always liked smalltalk's code browser that let you modify your methods while the program was running; this lead to some really fast development.
I'm wondering whether your experience w/ the hot-swap enabled debugger is the same: that most development is done while debugging. You just stub out your methods, start the program, and add code until it's running on your sample input. (then you sit down and refactor) Or does the system-programming nature of java come through, that you sit down and agonize for a week over which classes do what and where the methods go and what your invariants and edge cases are?
Man, I enjoyed smalltalk (digitalk smalltalk for the mac: ran in less than half a meg, including a GUI widget set)
Please comment on your experience w/ darcs; I can't quite figure out whether it is cool because it's good, or whether it's written in haskell.
they probably upload it to a server for processing.