I wonder if you've used OS X? I use it all the time and it never crashes. I normally reboot it every three weeks or so for a software update from apple.
When I first plugged in my digital camera I understood what "It just works" means. It recognized the camera the first time and with a single mouse click all my photos were downloaded.
Anyway, ease of use and stability are both hallmarks of Apple. I will concede that stability has only been true for a few months now.
I bet the resale value aspect of a mac changes soon. The hardware required to run OS X is so much higher than the previous versions that nobody in their right mind would pay nearly full price for an old mac. That old mac is slow.
That wasn't true when OS 9 was the OS of choice. Back then an old mac would do the job just fine.
In summary, now that Apple has joined the windows world of increasing hardware requirements you'll see the resale value of macs drop.
Vanguard (Written on my favorite notebook, a 500mhz iceBook)
When you pair a senior and a jr person together the jr person really does learn and improve. Also, the senior person does benefit from a second pair of eyes on the code.
XP doesn't say don't design the overall arch. It does say that you shouldn't attempt to design the every last detail. Those details are often implemented differently than the design because something new is discovered during development. Instead, XP suggests you design the easy 80% up front and leave the detailed 20% for the engineers.
Finally, studies prove that pairs take 58% of the time that two seperate people do. However, the lost 8% is more than recovered by the improved quality of code. (I'm too lazy to find the paper.)
Pairing up is good, small design up front is good (for smaller projects), and automated testing is good. Those are the three foundations of XP.
I'm fairly senior tech guy. I can sys admin and I can code. (I'm a better coder but that's besides the point).
I would take a guy in a call center. That job stinks. It's an entry level job that tries to serve senior level people. It's not a surprise that it doesn't work out well. The only way to fix it is to pay very well.
That model works in consulting. You give some comforts when you travel (the job is worse) but you get paid better. In support, you have to deal with more crap but if you get paid better you'll take the job.
Web services solve a business problem
on
Web Services
·
· Score: 2
The problem with CORBA, RMI and EJBs is that deploying them is a problem.
They are each nice is that they enable you to reuse code by tapping into them from a non-local box. For some of those above mentioned technologies, you can use them from different languages. However, they each require a client stub.
When you need to push a client stub around before you can access the system you have a deployment problem. With SOAP, you can consume those services right away without the need for a client stub.
I see solving the deployment problem as SOAP's chief advantage.
Vanguard
PS Now can anybody tell me why Gartner Group thinks that web services will take off inside the enterprise before it does across the web? Why use them internally when I don't have a deployment problem inside of my company?
Actually, the "super-node" idea was how they originally did SMP systems.
So each processor is connected by the front side bus. When you need a value you check each processor to see if it has it before you get it from main memory. Your alternative is to check with the controlling processor instead of each individual processor. Through the years they found that the loss of a processor just for admin tasks wasn't worth it. Now everybody shares admin load and everybody does work.
Because this admin load get's to be too much, they normally divide the machine into subsections. For example, the 64 CPU sun box I use at work is divided into 8 smaller sections. Each of the 8 CPUs are equally close to memory but they are far from the other "sub machine's" memory.
Anyway, I'm kind of rambling here but the general idea is that super computer builders have moved away from that idea in most of the models I've seen. Instead they try to keep the communication requirements low. (AKA, maximizing locality of reference)
If a "real" expert is around who knows of something different I'm all ears.
I'm not a real expert but I have recently taken a high performance computing course from somebody who is an expert for my comp sci masters.
The basic problem of adding more and more processors is keeping all the memory in sync. If you have a process that is running across 50 cpus the machine needs to ensure that if one of them updates a variable that all the others work with the current value. (Ok, it's more complicated than that but I'm not writing a book here)
The solution is to write your system so that the calculations can run as independently as possible. However, at 100 million processors it probably just doesn't fit the problem space.
Your company has it right. Intellij does it the same way. I downloaded it about three times before I bought it.
One difference, they don't let you use the same email twice. Something about setting up a fourth email account just to use the software for another 30 days (or whatever) made me realize that I was stealing. Now I own a copy and I'm happy to have paid for a good product.
Vanguard
PS This isn't meant to start an IDE war. Just an example of a good eval policy.
I run a rather unpopular website with some pictures of my coral reef tank. Sadly, Time Warner doesn't want me to run a server because of the traffic it could consume.
Now that they have a per for traffic model, can I run my server?
I've seen this so many times. Can somebody please explain what this quote is about? Did Bill G say it? I'm missing something. Everybody knows about this except me.
Actually, that's not true. A contract made that is nessessary for him to live his life is valid. For example, if he lived on his own and needed to sign an employment contract that contract would be binding. If he lives at home and signs up for a Columbia House cd contract he can bail.
Of course in this case, the contract would not be binding.
As for OS X being optimised for this kind of stuff, we are talking applications that nearly never call the OS for anything, so the impact of OS X is probably nil.
That doesn't sound right. Most unix systems, OSX included (and NT FYI) don't allow direct hardware calls. You can only access system resources through operating sytem APIs.
DOS is the only system I know of that lets you access the hardware directly. (I think NT let's you access graphics systems directly too. However, that has nothing to do with this test).
I could never figure this out. I'm not a master sys admin but I'm not a moron either. I'm a pretty good programmer and I have a recent master's degree in comp sci.
I suppose I didn't try *that* hard. However, I did try and in the end I bought a mac. I love OSX. Ok, I've gone off topic but my point is that if linux people care about market share (which they may not) then they need to fix the fonts. Really, it was my main reason for going over to apple.
When I wrote that I was thinking of all the memory related errors that are handled for you. In IT departments you get a developer base with a wide range of skill sets (read, some of us suck) so having things like automated garbage collection and memory allocation are nice to have.
I've given this a fair amount of thought and this is what I came up with:
When it's possible (right staff, right project, etc.) IT departments try to avoid C/C++.
Does that mean C is going away? No. It's also unfair to compare it to COBOL. When performance is important C/C++ is the only choice. However, if you have a chance, interpreted languages really do help with code quality and development time.
If you're coding for an IT department and you are doing it in C it's *probably* because they have no other choice.
Does this really do the job? I know nothing about wireless networks (even though mine works great) but I have to wonder if beefing up one end of the system really works.
I mean, if the card in my laptop still sucks can improving the base station really help? Don't you need to improve both sides for a system that sends and receives?
From the article I just did this testing today (2/14/2), but I'll post detailed specs on how to select a can, do the math, and build a waveguide antenna soon. Until then, consider buying the ARRL Antenna Handbook.
Read more??? Hey, I'm on slashdot. It's rare that I even read the linked article.:)
Slashdot Mirror
I wonder if you've used OS X? I use it all the time and it never crashes. I normally reboot it every three weeks or so for a software update from apple.
When I first plugged in my digital camera I understood what "It just works" means. It recognized the camera the first time and with a single mouse click all my photos were downloaded.
Anyway, ease of use and stability are both hallmarks of Apple. I will concede that stability has only been true for a few months now.
Vanguard
I bet the resale value aspect of a mac changes soon. The hardware required to run OS X is so much higher than the previous versions that nobody in their right mind would pay nearly full price for an old mac. That old mac is slow.
That wasn't true when OS 9 was the OS of choice. Back then an old mac would do the job just fine.
In summary, now that Apple has joined the windows world of increasing hardware requirements you'll see the resale value of macs drop.
Vanguard
(Written on my favorite notebook, a 500mhz iceBook)
I've done XP. It's a good thing.
When you pair a senior and a jr person together the jr person really does learn and improve. Also, the senior person does benefit from a second pair of eyes on the code.
XP doesn't say don't design the overall arch. It does say that you shouldn't attempt to design the every last detail. Those details are often implemented differently than the design because something new is discovered during development. Instead, XP suggests you design the easy 80% up front and leave the detailed 20% for the engineers.
Finally, studies prove that pairs take 58% of the time that two seperate people do. However, the lost 8% is more than recovered by the improved quality of code. (I'm too lazy to find the paper.)
Pairing up is good, small design up front is good (for smaller projects), and automated testing is good. Those are the three foundations of XP.
Vanguard
I'm fairly senior tech guy. I can sys admin and I can code. (I'm a better coder but that's besides the point).
I would take a guy in a call center. That job stinks. It's an entry level job that tries to serve senior level people. It's not a surprise that it doesn't work out well. The only way to fix it is to pay very well.
That model works in consulting. You give some comforts when you travel (the job is worse) but you get paid better. In support, you have to deal with more crap but if you get paid better you'll take the job.
The problem with CORBA, RMI and EJBs is that deploying them is a problem.
They are each nice is that they enable you to reuse code by tapping into them from a non-local box. For some of those above mentioned technologies, you can use them from different languages. However, they each require a client stub.
When you need to push a client stub around before you can access the system you have a deployment problem. With SOAP, you can consume those services right away without the need for a client stub.
I see solving the deployment problem as SOAP's chief advantage.
Vanguard
PS Now can anybody tell me why Gartner Group thinks that web services will take off inside the enterprise before it does across the web? Why use them internally when I don't have a deployment problem inside of my company?
Cool, my professor also worked at Cray but he is currently in IBM's Netfinity performace lab.
Anyway, no cache? I guess they must have some very fast ram if they don't need a local cache anymore. I'll check out the link.
Actually, the "super-node" idea was how they originally did SMP systems.
So each processor is connected by the front side bus. When you need a value you check each processor to see if it has it before you get it from main memory. Your alternative is to check with the controlling processor instead of each individual processor. Through the years they found that the loss of a processor just for admin tasks wasn't worth it. Now everybody shares admin load and everybody does work.
Because this admin load get's to be too much, they normally divide the machine into subsections. For example, the 64 CPU sun box I use at work is divided into 8 smaller sections. Each of the 8 CPUs are equally close to memory but they are far from the other "sub machine's" memory.
Anyway, I'm kind of rambling here but the general idea is that super computer builders have moved away from that idea in most of the models I've seen. Instead they try to keep the communication requirements low. (AKA, maximizing locality of reference)
If a "real" expert is around who knows of something different I'm all ears.
I'm not a real expert but I have recently taken a high performance computing course from somebody who is an expert for my comp sci masters.
The basic problem of adding more and more processors is keeping all the memory in sync. If you have a process that is running across 50 cpus the machine needs to ensure that if one of them updates a variable that all the others work with the current value. (Ok, it's more complicated than that but I'm not writing a book here)
The solution is to write your system so that the calculations can run as independently as possible. However, at 100 million processors it probably just doesn't fit the problem space.
Your company has it right. Intellij does it the same way. I downloaded it about three times before I bought it.
One difference, they don't let you use the same email twice. Something about setting up a fourth email account just to use the software for another 30 days (or whatever) made me realize that I was stealing. Now I own a copy and I'm happy to have paid for a good product.
Vanguard
PS This isn't meant to start an IDE war. Just an example of a good eval policy.
I run a rather unpopular website with some pictures of my coral reef tank. Sadly, Time Warner doesn't want me to run a server because of the traffic it could consume.
Now that they have a per for traffic model, can I run my server?
Yeah, I like it too.
I suspect it won't slow down the rate at which patches are accepted.
I was going to write the same thing. Let's leave the medical advice to doctors.
This link should work
I've seen this so many times. Can somebody please explain what this quote is about? Did Bill G say it? I'm missing something. Everybody knows about this except me.
Being a minor, NO contract he agrees to is valid
Actually, that's not true. A contract made that is nessessary for him to live his life is valid. For example, if he lived on his own and needed to sign an employment contract that contract would be binding. If he lives at home and signs up for a Columbia House cd contract he can bail.
Of course in this case, the contract would not be binding.
As for OS X being optimised for this kind of stuff, we are talking applications that nearly never call the OS for anything, so the impact of OS X is probably nil.
That doesn't sound right. Most unix systems, OSX included (and NT FYI) don't allow direct hardware calls. You can only access system resources through operating sytem APIs.
DOS is the only system I know of that lets you access the hardware directly. (I think NT let's you access graphics systems directly too. However, that has nothing to do with this test).
Vanguard
I could never figure this out. I'm not a master sys admin but I'm not a moron either. I'm a pretty good programmer and I have a recent master's degree in comp sci.
I suppose I didn't try *that* hard. However, I did try and in the end I bought a mac. I love OSX. Ok, I've gone off topic but my point is that if linux people care about market share (which they may not) then they need to fix the fonts. Really, it was my main reason for going over to apple.
It didn't look *that* big to me. I could be wrong but as I look at the picture that shows the cdrom, I see that it's large but not huge.
When I wrote that I was thinking of all the memory related errors that are handled for you. In IT departments you get a developer base with a wide range of skill sets (read, some of us suck) so having things like automated garbage collection and memory allocation are nice to have.
I've given this a fair amount of thought and this is what I came up with:
When it's possible (right staff, right project, etc.) IT departments try to avoid C/C++.
Does that mean C is going away? No. It's also unfair to compare it to COBOL. When performance is important C/C++ is the only choice. However, if you have a chance, interpreted languages really do help with code quality and development time.
If you're coding for an IT department and you are doing it in C it's *probably* because they have no other choice.
Does this really do the job? I know nothing about wireless networks (even though mine works great) but I have to wonder if beefing up one end of the system really works.
I mean, if the card in my laptop still sucks can improving the base station really help? Don't you need to improve both sides for a system that sends and receives?
From the article I just did this testing today (2/14/2), but I'll post detailed specs on how to select a can, do the math, and build a waveguide antenna soon. Until then, consider buying the ARRL Antenna Handbook.
:)
Read more??? Hey, I'm on slashdot. It's rare that I even read the linked article.
This article from extremetech is pretty good. The first half is mostly about the swing improvements. Part II (halfway down) covers the rest.
Can somebody explain why tivo's are $129 with directTv built in but $500+ on their own?
Thanks in advance,
Vanguard