You get some pretty interesting problems, when you increase the number of cores in your computer.
A couple of years ago, we replaced a 4-core IBM P5 with a 32-core HP DL 580. We tested it for a couple of months with just a user, or two, at a time. Then, we took a day and tested with the entire company (roughly 250 users). Thank goodness we did before we put it into production because, for some people, it was actually slower than the P5. It looked like it was going to be a disaster.
Fortunately, I had seen this problem before (on a Sequent Symmetry, of all things). I ran "strace" on the offending process, and sure enough, we were having problems with lock contention. We talked to our software vendor and, while it took a while for them to admit it was their problem (and probably cost us multiple thousands of dollars to have them fix it), they rewrote the code to use fewer locks. Problem solved.
Space 1999 was good when I was a kid, but when I re-watched it as an adult I found it terrible -- the show needed better writing
We could list that kind of series all day. The original battlestar galactica... Buck Rogers... The A team...
Yeah, but Battlestar Galactica had some decent episodes, even if some of the stories were stolen from movies. Space: 1999 (and Buck Rogers) redefined bad television. I've been re-watching it on DVDs from Netflix, and I'm up to the episode entitled "One Moment of Humanity". I have to say, it was possibly the worst hour of television ever produced. The "Star Wars Christmas Special" was better.
I'm up to the "Those of you who watched when it originally aired probably remember having nightmares after the "Dragon's Domain" episode (I certainly do), but so far, that's the best one I've seen. "Mission of the Darians" was tolerable, too.
They only introduced it on the 486? How did they do things like file locking and multi-processor synchronization on earlier architectures, without an instruction like that?
My 1200 baud modem is US Robotics. I took it apart one time and found it actually has two (6502) processors. I wonder if they use one for each direction, or one for the command processing, and one for the data.
I remember there was a thick coax variant of ethernet too (I think called 10 base 5).
Yes, and it was the original cabling used by Ethernet. Thin-net (10Base2) came later and it was a lot easier to work with than the thick stuff. I still have my 10Base2 crimping tool.
5.25" floppy ?? have not used one of these since 1995
I have a whole pile of stuff on 5.25" floppies. Fortunately, I saw the day coming, when they became extinct, so I tried to collect as many as I could, so I'd have spares. I think I managed to get four. Now if I could could just find a drive to read my 8" floppies, I'd be set.
Serial DB9 - I can still make these by hand! Definitely useful for many console RS232 equipment ports
Me too. On top of that, I'm the network administrator at my company, and a serial port is the best way to connect to an incorrectly configured network switch.
IDE Hard Drives - useable if you really had to, but why?
My main desktop at home is pre-SATA.
Modems (v21/v22)
I have a 1200baud and a few 14.4k modems at home, but nothing faster than that, and I haven't used them in forever. I used to live in one of the first markets to get DSL, and once you get high-speed Internet access, you kind of lose interest in modems.
I have an old Ohio Scientific machine at home (roughly contemporary with your Apple) that I used to screw around with once in a while, but I've found that, unless you need to connect a very specific piece of hardware to it (like an Apple II expansion card), it's easier to just write an emulator. Not only will your software run significantly faster, but machines that old generally had issues in the analog realm that don't exist with modern hardware. For instance, when you saved something to cassette on my OSI machine, you only had about an 80% chance of getting it back. With my emulator, I just need to point it to a file that looks like the OSI's cassette drive, and it reads it in perfectly, every time.
I'd be curious what some of these kids constitute as bullying, these days.
As a kid who was bullied in school, I'm not sure where I'd draw the line. I'd hesitate to anything less than physical contact bullying (including cyber-bullying), but I'd say anything that requires a hospital visit is definitely bullying, and needs to be addressed. I know people who disagree with the latter, and believe students should be allowed to work it out, on their own.
3 suggestions from someone who has worked for many nonprofits.
1) If you have any staff members, make technology a staff responsibility. More than just passing the buck, this puts control of tech decisions in the hands of the folks who work with it most often.
2) Ask Bob to chair the Board's nominating committee. This committee is responsible for finding new Board members (which you seem to need) and for monitoring Board terms and committee assignments. Hopefully, this will lead Bob to recognize that Board turnover is a healthy thing and that leaving one responsibility with one person for too long is not good for the organization.
3) Elect Bob as President or Chairperson of the Board. With so many other important things to focus on, he may be more willing to let go of his tech fiefdom.
are you prepared to defend the idea that increasing taxes on higher income individuals will somehow stimulate the economy?
Indirectly, yes. Lowering taxes on the middle class will stimulate the economy. The problem is that, unless we reduce the size of the government, we have to make up the difference, somewhere. We can't take any money from lower-income people, since they don't have any, so we have to get it from the wealthy.
Sure, it would be great if we could simply reduce the size of the government, but no administration in recent history (at least as far back as Eisenhower) has done that. Do that first. Then we can talk about lowering taxes on the wealthy, too.
Personally, I can't decide whether the biggest reason I can't stand them is the tiringly excessive force required to operate the keys or the deafening racket they produce.
Agreed. Keyboards should be quiet. Can you imagine an entire office full of Model M keyboards? Ow, my ears!
It would certainly be nice if they built them as solid as Model Ms today, but now they're so inexpensive, I'm not sure it matters that I have to buy a new one every few years, or so.
there still exists the challenge of getting past the barrier of infinite energy required to even match the speed of light
I've often wondered: if you're using a fuel where the energy produced is derived directly from the mass you input (like a fusion or matter-antimatter reaction), wouldn't the energy you get increase as you get closer to the speed of light and the mass of your fuel (assuming you're carrying it with you) increases? In other words, wouldn't the increase in energy cancel out the increase in mass, so your acceleration stays the same?
Of course, after doing some math, I wonder why anyone wants to go anywhere near the speed of light. Using standard Newtonian physics (without even taking relativity into account), I figured out that, assuming you're converting mass directly to energy with 90% efficiency, something like 95% of the mass of your ship would have to be fuel, and unless someone invents some type of anti-gravity technology, it would take over a year to accelerate to that velocity (and another year to decelerate at the end of your trip).
Oops, sorry to respond to myself, but I forgot to answer the question.
Started with Slackware (1.02?) - the one that was released in April of 1994 (I think).
Then, around Slackware 4, there were no updates for a very long time. I got tired of Red Hat people talking about cool new features that I couldn't use, so I switched.
After a few years of that, I started checking out other distributions and found that SuSE, with KDE, was really getting polished, so I started using that on my desktop, even though I stayed with Red Hat, on servers. Unfortunately, the last decent version of (Open)SuSE was 10.0 and, by that time, Gnome had significantly improved, so I switched again, to Fedora.
Currently, I'm still using Fedora 14 on my laptop, and I'd like to upgrade, but Gnome 3 is terrible. I can't switch back to OpenSuSE, because I'm not impressed with KDE 4 (though I haven't looked at it, in about 9 months), and I can't switch to Ubuntu, because I really hate Unity. I have no Idea where to go next.
We do, at my office, mainly because each one fills a slightly different need. We use Red Hat (with an up-to-date subscription) for the mission-critical servers, CentOS for the not-so-mission-critical servers, Fedora and Ubuntu for workstations (depending on user preference), and Debian and OpenBSD for older computers that would have been thrown in the trash, but are still usable for things like routers. (CentOS and Fedora refuse to install on a machine with less than 768Meg; Debian and OpenBSD work fine on a machine with 32Meg.)
Do we really want to cement ourselves with today's moed of thought?
If people would sit down and design them properly, it shouldn't be a problem. The Unix API was designed in the 70s and it still seems to work just fine.
Sorry to respond to myself, but as long as we're talking about programming, I can't help but wonder how much a real IDE would help. Whenever I talk to programmers who work on both Linux and Windows, they always tell me Windows programming is easier, because Linux doesn't have anything even remotely approaching Visual Studio. Is that what we need to get more desktop apps? (Yes, I've heard KDevelop is good. No, the programmers I've talked to don't think it's nearly as good as VS.)
YOU ASSHOLE I JUST WANT TO CHECK FACEBOOK NOT RECOMPILE A FUCKING OS.
Sorry, but I'm strongly inclined to believe you're the asshole. When you buy a Windows machine or Mac at the store, you're getting a machine that was designed, from the ground up, to run Windows or OSX. If you want proper hardware support, either make sure the machine you buy supports Linux, or buy a machine with Linux pre-installed. You have no trouble doing it for Windows or OSX. But no, it's a Linux problem...
Just out of curiosity, have you tried installing a generic copy of Windows on generic hardware? I have. It's not a pleasant experience.
Agreed. I shudder to think of all the time wasted creating TWO new DEs from scratch, and all the ways it could have been put to better use improving the existing ones.
how about getting rid of competing Gnome and KDE (and now Unity) desktops and agree on one standard desktop with a single API for everyone to write to
I'll half agree with you: I'd still like to see multiple desktops, but they should all have the same API. The Desktop should be the user's choice, not the programmer's. (The programmer shouldn't have to worry about it.)
Yes, you SHOULD worry about structuring your code, especially when you're learning.
Um... no. I know everyone likes to believe this, but it's baloney. When I'm learning a programming language, I like to produce something that works, in as short a time as possible. Having to worry about structure, on top of the syntax only increases that time. Pascal is like trying to teach first graders the alphabet, parts of speech, sentence structure, and symbolism, all at the same time. Sorry, but that's just not going to work.
I don't understand why you say it wasn't powerful enough for "experts." The Turbo/Borland/etc. Pascal compilers that people actually used were just as powerful as C compilers.
Well, it's been a while since I've done Turbo Pascal, but if I remember correctly, what made it powerful is that it didn't follow the Pascal standard. If that's true, I'd hardly call that powerful.
Slashdot Mirror
Of course not. It's closed-source software.
Thankfully, they were only reading the file. Why they were locking it, in the first place, I'll never know.
You get some pretty interesting problems, when you increase the number of cores in your computer.
A couple of years ago, we replaced a 4-core IBM P5 with a 32-core HP DL 580. We tested it for a couple of months with just a user, or two, at a time. Then, we took a day and tested with the entire company (roughly 250 users). Thank goodness we did before we put it into production because, for some people, it was actually slower than the P5. It looked like it was going to be a disaster.
Fortunately, I had seen this problem before (on a Sequent Symmetry, of all things). I ran "strace" on the offending process, and sure enough, we were having problems with lock contention. We talked to our software vendor and, while it took a while for them to admit it was their problem (and probably cost us multiple thousands of dollars to have them fix it), they rewrote the code to use fewer locks. Problem solved.
Yeah, but Battlestar Galactica had some decent episodes, even if some of the stories were stolen from movies. Space: 1999 (and Buck Rogers) redefined bad television. I've been re-watching it on DVDs from Netflix, and I'm up to the episode entitled "One Moment of Humanity". I have to say, it was possibly the worst hour of television ever produced. The "Star Wars Christmas Special" was better.
I'm up to the "Those of you who watched when it originally aired probably remember having nightmares after the "Dragon's Domain" episode (I certainly do), but so far, that's the best one I've seen. "Mission of the Darians" was tolerable, too.
Heck, I'd be happy modern windowing systems were better at handling the taskbar on the side on the screen, instead of on the top or bottom.
They only introduced it on the 486? How did they do things like file locking and multi-processor synchronization on earlier architectures, without an instruction like that?
My 1200 baud modem is US Robotics. I took it apart one time and found it actually has two (6502) processors. I wonder if they use one for each direction, or one for the command processing, and one for the data.
Yes, and it was the original cabling used by Ethernet. Thin-net (10Base2) came later and it was a lot easier to work with than the thick stuff. I still have my 10Base2 crimping tool.
I have a whole pile of stuff on 5.25" floppies. Fortunately, I saw the day coming, when they became extinct, so I tried to collect as many as I could, so I'd have spares. I think I managed to get four. Now if I could could just find a drive to read my 8" floppies, I'd be set.
Me too. On top of that, I'm the network administrator at my company, and a serial port is the best way to connect to an incorrectly configured network switch.
My main desktop at home is pre-SATA.
I have a 1200baud and a few 14.4k modems at home, but nothing faster than that, and I haven't used them in forever. I used to live in one of the first markets to get DSL, and once you get high-speed Internet access, you kind of lose interest in modems.
I have an old Ohio Scientific machine at home (roughly contemporary with your Apple) that I used to screw around with once in a while, but I've found that, unless you need to connect a very specific piece of hardware to it (like an Apple II expansion card), it's easier to just write an emulator. Not only will your software run significantly faster, but machines that old generally had issues in the analog realm that don't exist with modern hardware. For instance, when you saved something to cassette on my OSI machine, you only had about an 80% chance of getting it back. With my emulator, I just need to point it to a file that looks like the OSI's cassette drive, and it reads it in perfectly, every time.
As a kid who was bullied in school, I'm not sure where I'd draw the line. I'd hesitate to anything less than physical contact bullying (including cyber-bullying), but I'd say anything that requires a hospital visit is definitely bullying, and needs to be addressed. I know people who disagree with the latter, and believe students should be allowed to work it out, on their own.
3 suggestions from someone who has worked for many nonprofits.
1) If you have any staff members, make technology a staff responsibility. More than just passing the buck, this puts control of tech decisions in the hands of the folks who work with it most often.
2) Ask Bob to chair the Board's nominating committee. This committee is responsible for finding new Board members (which you seem to need) and for monitoring Board terms and committee assignments. Hopefully, this will lead Bob to recognize that Board turnover is a healthy thing and that leaving one responsibility with one person for too long is not good for the organization.
3) Elect Bob as President or Chairperson of the Board. With so many other important things to focus on, he may be more willing to let go of his tech fiefdom.
Indirectly, yes. Lowering taxes on the middle class will stimulate the economy. The problem is that, unless we reduce the size of the government, we have to make up the difference, somewhere. We can't take any money from lower-income people, since they don't have any, so we have to get it from the wealthy.
Sure, it would be great if we could simply reduce the size of the government, but no administration in recent history (at least as far back as Eisenhower) has done that. Do that first. Then we can talk about lowering taxes on the wealthy, too.
Agreed. Keyboards should be quiet. Can you imagine an entire office full of Model M keyboards? Ow, my ears!
It would certainly be nice if they built them as solid as Model Ms today, but now they're so inexpensive, I'm not sure it matters that I have to buy a new one every few years, or so.
Big Red what?
I've often wondered: if you're using a fuel where the energy produced is derived directly from the mass you input (like a fusion or matter-antimatter reaction), wouldn't the energy you get increase as you get closer to the speed of light and the mass of your fuel (assuming you're carrying it with you) increases? In other words, wouldn't the increase in energy cancel out the increase in mass, so your acceleration stays the same?
Of course, after doing some math, I wonder why anyone wants to go anywhere near the speed of light. Using standard Newtonian physics (without even taking relativity into account), I figured out that, assuming you're converting mass directly to energy with 90% efficiency, something like 95% of the mass of your ship would have to be fuel, and unless someone invents some type of anti-gravity technology, it would take over a year to accelerate to that velocity (and another year to decelerate at the end of your trip).
Oops, sorry to respond to myself, but I forgot to answer the question.
Started with Slackware (1.02?) - the one that was released in April of 1994 (I think).
Then, around Slackware 4, there were no updates for a very long time. I got tired of Red Hat people talking about cool new features that I couldn't use, so I switched.
After a few years of that, I started checking out other distributions and found that SuSE, with KDE, was really getting polished, so I started using that on my desktop, even though I stayed with Red Hat, on servers. Unfortunately, the last decent version of (Open)SuSE was 10.0 and, by that time, Gnome had significantly improved, so I switched again, to Fedora.
Currently, I'm still using Fedora 14 on my laptop, and I'd like to upgrade, but Gnome 3 is terrible. I can't switch back to OpenSuSE, because I'm not impressed with KDE 4 (though I haven't looked at it, in about 9 months), and I can't switch to Ubuntu, because I really hate Unity. I have no Idea where to go next.
Not if you add EPEL:
http://fedoraproject.org/wiki/EPEL
We do, at my office, mainly because each one fills a slightly different need. We use Red Hat (with an up-to-date subscription) for the mission-critical servers, CentOS for the not-so-mission-critical servers, Fedora and Ubuntu for workstations (depending on user preference), and Debian and OpenBSD for older computers that would have been thrown in the trash, but are still usable for things like routers. (CentOS and Fedora refuse to install on a machine with less than 768Meg; Debian and OpenBSD work fine on a machine with 32Meg.)
So what you're basically saying is: "Works for me".
If people would sit down and design them properly, it shouldn't be a problem. The Unix API was designed in the 70s and it still seems to work just fine.
Sorry to respond to myself, but as long as we're talking about programming, I can't help but wonder how much a real IDE would help. Whenever I talk to programmers who work on both Linux and Windows, they always tell me Windows programming is easier, because Linux doesn't have anything even remotely approaching Visual Studio. Is that what we need to get more desktop apps? (Yes, I've heard KDevelop is good. No, the programmers I've talked to don't think it's nearly as good as VS.)
Sorry, but I'm strongly inclined to believe you're the asshole. When you buy a Windows machine or Mac at the store, you're getting a machine that was designed, from the ground up, to run Windows or OSX. If you want proper hardware support, either make sure the machine you buy supports Linux, or buy a machine with Linux pre-installed. You have no trouble doing it for Windows or OSX. But no, it's a Linux problem...
Just out of curiosity, have you tried installing a generic copy of Windows on generic hardware? I have. It's not a pleasant experience.
Agreed. I shudder to think of all the time wasted creating TWO new DEs from scratch, and all the ways it could have been put to better use improving the existing ones.
I'll half agree with you: I'd still like to see multiple desktops, but they should all have the same API. The Desktop should be the user's choice, not the programmer's. (The programmer shouldn't have to worry about it.)
Um... no. I know everyone likes to believe this, but it's baloney. When I'm learning a programming language, I like to produce something that works, in as short a time as possible. Having to worry about structure, on top of the syntax only increases that time. Pascal is like trying to teach first graders the alphabet, parts of speech, sentence structure, and symbolism, all at the same time. Sorry, but that's just not going to work.
Well, it's been a while since I've done Turbo Pascal, but if I remember correctly, what made it powerful is that it didn't follow the Pascal standard. If that's true, I'd hardly call that powerful.