"That's because there are tradeoffs in everything... if you've been told to "clean up the codebase", take a bit to look at the codebase, and tell your manager that it's going to take X amount of time to do that the manager has to decide whether or not it's worth the time to do so -- since otherwise your time could be spent doing other things. And odds are the cleaning up isn't going to show an immediate return to the company. Of course, there are other plusses to cleaning up code -- like doing it right may mean that you can implement future features in less time -- but those are harder to quantify."
So what you're saying is:
Harder to quantify = "Manager is without a clue about the technology." Immediate return to the company = "We can always fix it later. Make money now, I'll be on to some other project and you'll be here cleaning up the mess."
Yep, sounds like the complete CRAP work being done by most of the software industry. At least you finally put the blame where it belongs in most of the cases. Clueless, planless managers who aren't able to even fathom complex software updates their companies need to survive.
They make pretty Gantt charts though -- that get longer and longer and longer...
Would you also possibly have a high tolerance for "social" pain?
Could this be a future extension of this study to find whether or not people who push themselves physically to the limits of human performance also appear to be the people who are the very least concerned with what others think about them?
Most "great" athletes say they are competing against only one person -- themselves.
In addition, could this relate at a low level to people with high tolerance for pain coming across as somewhat "anti-social" because they have little or no reaction to the "social pain" others may feel deeply when rejected or ignored?
Ahh... yep, old info on my part. Oh well. Still stops at m for a reason though -- I think your 10-year-old server (BIND 4?) is the limitation. Otherwise they would have just extended the roots out to z instead of splitting the GTLD servers off, methinks. Unless that was politically oriented (always a bad design choice).
I never mentioned Verisign's wildcard at all - I was just explaining to the previous poster that said that all queries go to the root that they really shouldn't. If things are set up correctly.
The roots are not the GTLD servers. GTLD is the delegation of a number of TLD's down a few years ago to take some load off the root servers.
(a.root-servers.net - f.root-servers.net are the roots, a.gtld-servers.net - f.gtld-servers.net are the gtld servers -- why a through f? there are old named's out there still to this day that can't handle answers larger than a certain size. "f" is the limit to the number of responses they can deal with.)
Otherwise, you're close to being correct, but you forgot that every system along the way back to the roots also typically has a cache of information and this is what TTL times and other times in the SOA record are for. How long to keep the cached information.
If you're going to a well-travelled website you're not hitting the roots, the gtld's, or anything above your ISP's DNS server if anyone else has gone there before you did within the allotted time period for TTL caching in the zone. You're getting a non-authoritive answer from your ISP's server.
All this "oh, the firewalls and logging servers overload the roots" is just FUD and misunderstanding how DNS really works. If people keep their TTL times high except prior to making changes, the caches can do their jobs. (Some servers insert an arbitrary minimum TTL for zones that are abusing this.)
If you're just looking to work with a small microprocessor/computer to get that nostalgic feel, just take a modern microcontroller and start breadboarding.
Microcontroller products from Microchip and Atmel fit the bill nicely.
For support and help with Microchip PIC microcontroller development, you can hardly go wrong with the MIT-PICList, and for Atmel AVR micros AVRFreaks is the place to go.
Then you'll probably find yourself over at DigiKey buying parts for your projects after you have gained some insight into just how cool it is to have a 40 MHz processor with 2K or more of FLASH RAM on-board in a 18-pin device right at your fingertips that only needs a PC and some imagination to program to act like just about any logic device.
When the project is started, document all promises, timelines and what's going to happen (what you're going to do).
Waffling? Show the document. Ask why it needs to be changed. Call a meeting and CHANGE the document if necessary in front of everyone. "Mr. Boss, I've called this meeting to make changes to the requirements of the system everyone agreed to, and I just want to make sure everyone understands the changes you've requested. Could you detail those for us?"
Changing direction. With the document in hand, ask why? CHANGE the document. Same as above.
Act like he knew nothing about it? Show the document.
Bottom line. Be smarter than he is. He's not your enemy -- you are on the same team -- but if he's making your job more difficult instead of better, no one knows unless you've documented it along the way. In full view and with the approval of your internal "customers".
Sorry, I got the impression you were one of the guys here thinking that because something looks like their "walkie-talkies" they had when they were 10 years old, they think there's no technology behind it.
A bunch of the new APCO-25 digital radios for public safety agencies look a lot like those kid's toys too, but they contain technology far beyond what the average Slashdot poster could figure out.
My apologies to you, and hope you have a nice day!
Yeah, I'm going to get modded down as flaimbait... because the world full of electronics "consumers" who couldn't resolder a bad wire in their precious PDA themselves -- will be modding.
But...
I'll take these words back when you've designed and marketed a modern multi-band HF/VHF/UHF radio the size of a small car stereo that puts out 100 watts of RF and has a built in autotuner to tune an external motorized antenna automatically.
Or you could just go down to your store and buy one from Yaesu.
Let's make it even easier. You build a K2 *KIT* that's already designed for you, in your spare time, from the folks at http://www.elecraft.com/ and then come back and tell me ham radio's "low tech".
They probably also don't need a computer for whatever it is that they're doing. Pencil, paper, and a filing cabinet are probably faster for these people.
Let them be intimidated. Who cares?
For those willing to put in a minimal amount of effort and learn how to use the machine (key word... machine...) they'll reap the benefits if there are any to be had in their particular endeavour.
For those willing to learn, let them pay those willing to teach an appropriate wage also.
And for those of us who just get things done with our computers, let's stop worrying about if Aunt Tillie likes the bloody thing. It's a box with a display, a keyboard, some storage devices, and some wires to her, and that's all it needs to be.
You don't quite get it do you? If the driver is useful and works, Linux won't "change" because there would be pressure on the kernel folks making that change not to do it, so the people they're serving could continue to use the driver.
Linux and Free Software are about service. Some of the best customer service in the world. "We'll try to make what you want, but if we don't quite get it right, here's the source code, you can work on it yourself with whatever resources you might bring to the table. Enjoy."
Liability. Sheesh. What brain-dead suit came up with that one? "Those linux guys are going to change everything and my precious little 200 line C code driver won't work -- but I refuse to release it to the them fix their own problem so I can continue to sell my product -- which is HARDWARE."
TI and the manufacturers using their chipset are making a big mistake. But TI's been getting their butts kicked for years by numerous other chip vendors. They even had to buy Burr-Brown to have decent analog devices.
It's all cyclical though -- maybe they'll figure it out before they go under. Maybe they won't.
Because my boss demands the use of RedHat on servers when Debian would certainly be a much better choice in terms of managing the complexity -- a few apt-get commands and everything's the same. AND the old Sparc boxes could run it... but not RedHat... so we manage Linux/Solaris instead of Linux/Linux.
Plus, Debian "stable" is just that. STABLE. Rock solid. It's always there. Knowing that RedHat tries to hard to stay on the bleeding edge makes package upgrades worrisome. In fact, they've had this anti-upgrade effect on my boss... who chose RedHat to begin with. Hmm... interesting how that decision is messing with things again...
On RedHat, you either pay for RedHat Network, install apt and hope some badly created package doesn't blow up things while using it, install Red Carpet and really screw up the machine, or just don't upgrade. ALL of which are silly in light of the availability of a strict package system like Debian's that WORKS 100% of the time when using "stable".
But... the bosses see RedHat on the shelves and in the news.
So guess what every Linux-based solution has always been used at the companies I've worked for?
When presented with the question "Why?" the answer is usually "because there's commercial support", and "because we've already made that decision" (without any technical forethought, I might add...), but I've not YET seen a company I've worked for EVER pay for commercial support from RedHat -- it's a B.S. response. Considering that one of my employers blew over $100 million in a couple of years time, they definitely had the cash to take advantage of RedHat's advanced services.
The real reason? Marketing. The same reason people use M$ products over better ones. RedHat has them by the eyeballs with the fancy logo and the marketing-speak, and the brain's not engaged. Like moths to a fire...
You've obviously never had the joy of listening to a 3 year old hard disk that lived for three years without anything but convective cooling whining -- "I'm losing my bearings, you really should replace me." song from an iMac.
And the other (harder and more geeky) contests/events are ignored.
Here's a good one... http://www.arrl.org/contests/rules/2003/10 -ghz.htm l [10 GHz and up contest] - try doing THAT with your Pringles Can antenna (hint: Leave the pringles can at home in the garbage and get a real dish.)
Or maybe a little more realistic...
http://www.arrl.org/contests/rules/2003/june-vhf .h tml [June VHF QSO Party] - put a whole crapload of directional antennas up on bands that are supposed to be "local" communications bands and then talk hundreds of miles away on them.
The highly technical contests are in need of more geeks in most areas of the country. Field Day is fun, but not nearly as challenging (unless you keep spilling your beer) as the VHF and up contests that never make Slashdot.
Satellites, super-low-power repetitive digital modes (Amateurs are confirming moon-bounce "satellite" contacts with minimal gear on UHF and EXCELLENT DSP processing in PC's now), and the highly technical stuff never gets the accolades or the press -- but it's all WAY better than Field Day.
You take your standard FM VHF/UHF/Whatever voice 2-way bandwidth of 5 KHz of bandwidth modulated for voice peaks at around 3.5/4.0 KHz.
You stuff that into a sound card and a codec like ADPCM (32Kb/sec) or GSM (8-11Kb/s). When you digitize it you do it at a typical rate of 44.1 KHz (typical sound card sampling rate) and THEN stuff it through the CODEC making it much much smaller.
You pop it out the other side from another PC and sound card, and put it right back into the same 5KHz channel size as before.
Do you think every cell phone out there uses 200KHz wide channels?
Re:Hmmmmmm I wonder...
on
Working Hard?
·
· Score: 1
You know anyone who's worked for a company seven years in the U.S.? I'm impressed. That's not the norm.
Slashdot Mirror
Dispute the (automatic) credit card charge. You didn't get what you paid for.
"That's because there are tradeoffs in everything... if you've been told to "clean up the codebase", take a bit to look at the codebase, and tell your manager that it's going to take X amount of time to do that the manager has to decide whether or not it's worth the time to do so -- since otherwise your time could be spent doing other things. And odds are the cleaning up isn't going to show an immediate return to the company. Of course, there are other plusses to cleaning up code -- like doing it right may mean that you can implement future features in less time -- but those are harder to quantify."
:
So what you're saying is
Harder to quantify = "Manager is without a clue about the technology."
Immediate return to the company = "We can always fix it later. Make money now, I'll be on to some other project and you'll be here cleaning up the mess."
Yep, sounds like the complete CRAP work being done by most of the software industry. At least you finally put the blame where it belongs in most of the cases. Clueless, planless managers who aren't able to even fathom complex software updates their companies need to survive.
They make pretty Gantt charts though -- that get longer and longer and longer...
Would you also possibly have a high tolerance for "social" pain?
Could this be a future extension of this study to find whether or not people who push themselves physically to the limits of human performance also appear to be the people who are the very least concerned with what others think about them?
Most "great" athletes say they are competing against only one person -- themselves.
In addition, could this relate at a low level to people with high tolerance for pain coming across as somewhat "anti-social" because they have little or no reaction to the "social pain" others may feel deeply when rejected or ignored?
My God...
Mozilla's morphing into Xemacs!
Ahh... yep, old info on my part. Oh well. Still stops at m for a reason though -- I think your 10-year-old server (BIND 4?) is the limitation. Otherwise they would have just extended the roots out to z instead of splitting the GTLD servers off, methinks. Unless that was politically oriented (always a bad design choice).
I never mentioned Verisign's wildcard at all - I was just explaining to the previous poster that said that all queries go to the root that they really shouldn't. If things are set up correctly.
The roots are not the GTLD servers. GTLD is the delegation of a number of TLD's down a few years ago to take some load off the root servers.
(a.root-servers.net - f.root-servers.net are the roots, a.gtld-servers.net - f.gtld-servers.net are the gtld servers -- why a through f? there are old named's out there still to this day that can't handle answers larger than a certain size. "f" is the limit to the number of responses they can deal with.)
Otherwise, you're close to being correct, but you forgot that every system along the way back to the roots also typically has a cache of information and this is what TTL times and other times in the SOA record are for. How long to keep the cached information.
If you're going to a well-travelled website you're not hitting the roots, the gtld's, or anything above your ISP's DNS server if anyone else has gone there before you did within the allotted time period for TTL caching in the zone. You're getting a non-authoritive answer from your ISP's server.
All this "oh, the firewalls and logging servers overload the roots" is just FUD and misunderstanding how DNS really works. If people keep their TTL times high except prior to making changes, the caches can do their jobs. (Some servers insert an arbitrary minimum TTL for zones that are abusing this.)
You sound like you need the serivces of Terry Tate, Office Linebacker...
"You kill the Joe, you make some Mo' Baby! Whoo whooooo!"
You know how it goes... I trust you, but my company can't. :-)
I keep hearing 'bout these disgruntled developers. I'm waiting to meet a gruntled one.
Yeah right... we're all going to install binaries from some jackass we only have an AIM account for.
What are you, new?
If you're just looking to work with a small microprocessor/computer to get that nostalgic feel, just take a modern microcontroller and start breadboarding.
Microcontroller products from Microchip and Atmel fit the bill nicely.
For support and help with Microchip PIC microcontroller development, you can hardly go wrong with the MIT-PICList, and for Atmel AVR micros AVRFreaks is the place to go.
Then you'll probably find yourself over at DigiKey buying parts for your projects after you have gained some insight into just how cool it is to have a 40 MHz processor with 2K or more of FLASH RAM on-board in a 18-pin device right at your fingertips that only needs a PC and some imagination to program to act like just about any logic device.
Have fun twiddling those bits, boys and girls.
I have moderator points right now, but there's nothing called "Idiotic Posting of something you already linked to" in the checkbox!
;-)
The link and an excerpt would have been plenty.
Of course, moderators can waste their points modding this reply as "Offtopic" as they'd like, but that's blatently obvious.
It's a competition to move large trucks autonomously. Did you read the competition information? It's sponsored by the U.S. Army.
If the competition were to move a vehicle of any type across the desert you could build an autonomous small aircraft and fly it. That's not the point.
But there's nothing stopping you from trying -- you don't have to compete against anyone but yourself.
So what you're saying is we were once ahead, but now we're falling behind because we were ahead... :-)
Hell, not only does the disk itself arrive but the guy to put it in. NetApp sells and SUPPORTS a great product.
This is a relatively easy situation to fix.
Document. Before work begins.
When the project is started, document all promises, timelines and what's going to happen (what you're going to do).
Waffling? Show the document. Ask why it needs to be changed. Call a meeting and CHANGE the document if necessary in front of everyone. "Mr. Boss, I've called this meeting to make changes to the requirements of the system everyone agreed to, and I just want to make sure everyone understands the changes you've requested. Could you detail those for us?"
Changing direction. With the document in hand, ask why? CHANGE the document. Same as above.
Act like he knew nothing about it? Show the document.
Bottom line. Be smarter than he is. He's not your enemy -- you are on the same team -- but if he's making your job more difficult instead of better, no one knows unless you've documented it along the way. In full view and with the approval of your internal "customers".
Try it. It'll work.
And some of us run sun4u/linux. :-) Old Ultra 1's make pretty good commodity Linux boxen.
Sorry, I got the impression you were one of the guys here thinking that because something looks like their "walkie-talkies" they had when they were 10 years old, they think there's no technology behind it.
A bunch of the new APCO-25 digital radios for public safety agencies look a lot like those kid's toys too, but they contain technology far beyond what the average Slashdot poster could figure out.
My apologies to you, and hope you have a nice day!
Low tech, my ass!
...
Yeah, I'm going to get modded down as flaimbait... because the world full of electronics "consumers" who couldn't resolder a bad wire in their precious PDA themselves -- will be modding.
But
I'll take these words back when you've designed and marketed a modern multi-band HF/VHF/UHF radio the size of a small car stereo that puts out 100 watts of RF and has a built in autotuner to tune an external motorized antenna automatically.
Or you could just go down to your store and buy one from Yaesu.
Let's make it even easier. You build a K2 *KIT* that's already designed for you, in your spare time, from the folks at http://www.elecraft.com/ and then come back and tell me ham radio's "low tech".
73 OM DE WY0X
They probably also don't need a computer for whatever it is that they're doing. Pencil, paper, and a filing cabinet are probably faster for these people.
Let them be intimidated. Who cares?
For those willing to put in a minimal amount of effort and learn how to use the machine (key word... machine...) they'll reap the benefits if there are any to be had in their particular endeavour.
For those willing to learn, let them pay those willing to teach an appropriate wage also.
And for those of us who just get things done with our computers, let's stop worrying about if Aunt Tillie likes the bloody thing. It's a box with a display, a keyboard, some storage devices, and some wires to her, and that's all it needs to be.
You don't quite get it do you? If the driver is useful and works, Linux won't "change" because there would be pressure on the kernel folks making that change not to do it, so the people they're serving could continue to use the driver.
Linux and Free Software are about service. Some of the best customer service in the world. "We'll try to make what you want, but if we don't quite get it right, here's the source code, you can work on it yourself with whatever resources you might bring to the table. Enjoy."
Liability. Sheesh. What brain-dead suit came up with that one? "Those linux guys are going to change everything and my precious little 200 line C code driver won't work -- but I refuse to release it to the them fix their own problem so I can continue to sell my product -- which is HARDWARE."
TI and the manufacturers using their chipset are making a big mistake. But TI's been getting their butts kicked for years by numerous other chip vendors. They even had to buy Burr-Brown to have decent analog devices.
It's all cyclical though -- maybe they'll figure it out before they go under. Maybe they won't.
Because my boss demands the use of RedHat on servers when Debian would certainly be a much better choice in terms of managing the complexity -- a few apt-get commands and everything's the same. AND the old Sparc boxes could run it... but not RedHat... so we manage Linux/Solaris instead of Linux/Linux.
Plus, Debian "stable" is just that. STABLE. Rock solid. It's always there. Knowing that RedHat tries to hard to stay on the bleeding edge makes package upgrades worrisome. In fact, they've had this anti-upgrade effect on my boss... who chose RedHat to begin with. Hmm... interesting how that decision is messing with things again...
On RedHat, you either pay for RedHat Network, install apt and hope some badly created package doesn't blow up things while using it, install Red Carpet and really screw up the machine, or just don't upgrade. ALL of which are silly in light of the availability of a strict package system like Debian's that WORKS 100% of the time when using "stable".
But... the bosses see RedHat on the shelves and in the news.
So guess what every Linux-based solution has always been used at the companies I've worked for?
When presented with the question "Why?" the answer is usually "because there's commercial support", and "because we've already made that decision" (without any technical forethought, I might add...), but I've not YET seen a company I've worked for EVER pay for commercial support from RedHat -- it's a B.S. response. Considering that one of my employers blew over $100 million in a couple of years time, they definitely had the cash to take advantage of RedHat's advanced services.
The real reason? Marketing. The same reason people use M$ products over better ones. RedHat has them by the eyeballs with the fancy logo and the marketing-speak, and the brain's not engaged. Like moths to a fire...
You've obviously never had the joy of listening to a 3 year old hard disk that lived for three years without anything but convective cooling whining -- "I'm losing my bearings, you really should replace me." song from an iMac.
And the other (harder and more geeky) contests/events are ignored.
0 -ghz.htm l [10 GHz and up contest] - try doing THAT with your Pringles Can antenna (hint: Leave the pringles can at home in the garbage and get a real dish.)
f .h tml [June VHF QSO Party] - put a whole crapload of directional antennas up on bands that are supposed to be "local" communications bands and then talk hundreds of miles away on them.
Here's a good one...
http://www.arrl.org/contests/rules/2003/1
Or maybe a little more realistic...
http://www.arrl.org/contests/rules/2003/june-vh
The highly technical contests are in need of more geeks in most areas of the country. Field Day is fun, but not nearly as challenging (unless you keep spilling your beer) as the VHF and up contests that never make Slashdot.
Satellites, super-low-power repetitive digital modes (Amateurs are confirming moon-bounce "satellite" contacts with minimal gear on UHF and EXCELLENT DSP processing in PC's now), and the highly technical stuff never gets the accolades or the press -- but it's all WAY better than Field Day.
73 DE WY0X
No no no...
You take your standard FM VHF/UHF/Whatever voice 2-way bandwidth of 5 KHz of bandwidth modulated for voice peaks at around 3.5/4.0 KHz.
You stuff that into a sound card and a codec like ADPCM (32Kb/sec) or GSM (8-11Kb/s). When you digitize it you do it at a typical rate of 44.1 KHz (typical sound card sampling rate) and THEN stuff it through the CODEC making it much much smaller.
You pop it out the other side from another PC and sound card, and put it right back into the same 5KHz channel size as before.
Do you think every cell phone out there uses 200KHz wide channels?
You know anyone who's worked for a company seven years in the U.S.? I'm impressed. That's not the norm.