Now go back and look at what a Sun 10000 was doing in about 1990. If I recall it had dual 128 bit busses, each running at 50 Mhz. I am not talking about your punny desktop, small server boxes, I am talking about your large 60+ CPU server systems.
These things cost money so they don't make it into comodity hardware, but when you need the system throughput Sun is actually a pretty good company to go with.
Now just add wireless, and the ability to do H.323 and you can call me Dick T.
Seriously, this is cool. Just bought my 2 year old a $20 dollar digital camera to let her play with, not a mega pixel camera either, but good enough for her to play with and it is cheaper than film.
I love hearing that people are finally starting to publish intentions. I have been hearing rumors about this for a year or so now, since an EVP where I worked started talking about plugging a Fibre into the side of the microprocessor (and he wanted to own that connection)
As is normal, he missed completely thinking it would be a 10GbE fiber for networking, rather than a 40+GB connection to main memory...
The comments on working on the I/O side of the processor were right on (I read the EETimes article, rather than the Register article to get "real" facts ). For years Sun was known for having the slowest RISC processor in the business, however they had the fastest boxes. No one seemed to understand this, until they realized that they were running multiple 128 bit memory buses at rather good clock rates. That was better than 10 years ago, and just now we are starting to see memory busses approaching this level in their competitors hardware.
So they are saying that if you take out a large telephone networks Central Office, people connected to this office will suffer lost connections. Infact some long distance connectivity will suffer as well.
Why does this suprise anyone. Hmmm let me see, if you take out your ISP, all of the sudden you will loose connectivity to the internet unless you pay A LOT of money to have a second line put in. Even then the chance that both of those lines run through some common area is pretty high.
Things are easy to engineer with fully redundancy, what isn't easy is to do it cheaply enough that people will still be willing to pay for it.
Of course I send them to my representatives. In the state of Oregon, I am stuck with Sen. Wyden (D- Ore) and Sen. Smith (R- Ore)and am represented by Rep David Wu (D- Ore) in the house. I write one of them at least every month on topics ranging from the environment, SPAM, technology issues, and many others.
Every time I have gotten a response from one of them. Why would you expect a response from a representative of another area anyway ? It is not their job to represent your views in congress, but those of their constituents.
Well it looks like we will never be able to download music on the internet. First it was because we weren't paying the artists, just shipping the music to each other, now it is because we have exclusive contracts with the artists to only use the one service that pays them...
I Want my Net-TV (to paraphrase Sting from a Dire Straights song)
I'd like to know where you can get a T-1 for $70 a month, if ya do, tell me where. All the Mobilstar Starbucks locations had T-1s rather than DSL or something cheaper Sounds like a business problem to me. No wonder they were loosing money. As I said, I think you would have to sell this service to a business for under a few hundred dollars a month. Do you think starbucks cares what their ping times are, DSL is fine...
Obviously we are not paying "costs" if people are going bankrupt providing the service. I can understand if it were one maybe two badly run companies, however the whole industry is quickly going bankrupt...
You can't tell me that a $200 dollar access point, and a $70/month Broadband connection to the internet is a barrier to entry... That is all it takes to connect a starbucks to the internet. I'll start a business now, except not enough people would be willing to pay me the $200-$300/month it would cost to make this profitable enought to be interesting. (I need to make a few tens of thousand a month to pay my staff to run it, plus clear a profit for myself)
2 years ago, it was get a customer base, then figure out how to exploit them to make a profit. Now people have realized that there are no barriers to entry, so you can't raise prices later. You have to show up front how you can make money off of each and every customer from day one, then hope you get enough of them to overcome useless overhead in the corporation (read, the CEO, CTO, C??, and much of the marketing dept.)
The days of giving away dollar bills for 3 quarters to generate revenue are over for the internet, show me how you can make money, or go the way of the Dodo bird
This is what most large companies do, buy a bunch of leased lines and run an ATM backbone (or other technology) over it to provide an internal WAN capability...
Why not do it over the internet in general ??? Well this way you have guaranteed bandwidth characteristics, as much data hiding as you want, and each office does not have to expose itself to the internet in general
One thing Microsoft does is to make sure that drivers, etc. are generally backwards compatible, sometimes to it's detriment. I can tell you haven't written drivers for Microsoft... There are VERY large differences between 95, 95 OSR2.1 (that was the last version shipped) 98, SE, ME, and the various NTs...
The big differences are between SE/ME 9X/NT4, however there are subtle differences between NT4 and W2K, and even some fun differences between the various Service Packs for NT4. Don't even get me started with NT versions prior to 4.0.
Of course this took a little bit of effort to read the linux kernel mail list. Interesting start the first paragraph is
The chances are that you've seen an interview or two, in print or on television, with the fellow who a decade ago had the sheer audacity to sit down and start writing his own version of Unix. If you have, you've probably been impressed with a polite and self-assured young man who was probably giving rote answers to the same questions that are asked in every interview.
Hmmm... seems to backup my point, maybe it is a time for a slash interview and see what the/. crowd can come up with
Actually more humourously, hack the phone to return a GPS coordinate of you standing RIGHT BEHIND THEM...
Oh wait, we aren't watching Real Genius are we, never mind
Actually I did this in the mid 80s to keep my aging PC working...
Install memory extender to allow 1 MB RAM
Boot off of floppy A (Did I say REALLY old PC - Before hard drives even)
Install EMS memory handler to get at memory
Install RAM DISK of size 512 KB
Copy contents of floppy B onto RAM disk (basically Borland Turbo C 1.x)
Replace floppy B with modem software
Call campus network and connect to VAX systems
Leave computer on in dorm room for months at a time.
Basically this gave me VERY fast access to a compiler (remember when you could fit Turbo C onto a single floppy), and room for temp storage as well...
Problems with stupid journalists
on
Torvalds Tells All
·
· Score: 2, Insightful
When a journalist can't come up with an origional question, they seem to want to come up with a question that they know might generate controversy if it is answered, or not answered.
Do you have a beef with RMS over GNU/Linux ?
Do we have ground troops in Afghanistan ?
Have you had sexual relations with an Intern ?
When will journalists learn to at least ask a good question.
What do you think RMSs biggest contribution to the Open source movement is ?
Who inspires you today, who do you see as your idol ?
What message would you like to deliver to todays incoming college Computer Science Freshmen, what do you think they should be looking at ?
What are your choices... I mean MS upgrades their OS on a yearly bassis (to almost no effect) first it was Win 95 (and a OEM only service release every year after that) then Win 98, Win 98 SE, Win ME, Win XP... Seems pretty frequent to me...
As for not being able to upgrade because your binaries only work with a given version... Well that is why you either get source, or go with a company that will keep its software updated (and probabaly pay maintenence on the software as well)
I was glad to see that development on the 2.5 kernel will begin within the month. I know it is VERY important to get the stable kernels as stable as possible before you release the hordes on the next project. I think Linus deserves praise for keeping the 2.5 kernel back until he has as close to perfection as he can get on the 2.4 kernel.
Once he is done there, it is time to start adding features again... Too bad microsoft only sends out a new service pack about every 6-12 months vs. the 10 kernel updates that have happened since the 2.4 kernel came out last year, maybe if they could do that they might be able to put an IIS server on the internet without it being hacked from here to the far reaches of space.
Maybe it is time to download 2.4.11 and get my kernel updated from 2.4.4
Well here I thought we were staying with commodity hardware, but if we are going to go beyond commodity hardware into stuff that is engineered to provide low latency, high bandwidth, lets look at technologies like Infiniband which is engineered to avoid almost all OS latencies by delivering data directly to applications from hardware with OS bypass (almost no software between your app and the hardware anymore), and at 2 Gb/sec it isn't too bad on the performance scale. However I am betting that when this stuff comes out it will be a little pricey and go slightly over the 210,000 dollar budget
Absolutely none I would hope... The dataset resides on a centrally managed server, and because they are running a Linux desktop I get to laugh at what a trojan horse virii can do to a user account on a Unix box. This can also be removed as a problem by putting a keyboard, mouse and monitor on the desktop and locking the PC into a cabnet under each desk... What the user can't touch the user can't screw-up
That is a serious problem though, and one I assume Beowulf clusters will take care of, what if a node goes down in the middle of processing, how does the cluster respond to it ?
I have not had the chance to play with Beowulf clusters at all. Do I still get a local desktop on certain clustered computers ???
The ultimate Linux selling tool, every linux box in your company is a node in a cluster, add a few servers for extra speed, add a few computers to provide file I/O and backup capability, and you have one of the fastest supercomputers available to your company without having to spend an extra dime (everyone needs a desktop anyway). Can you imagine the extra cycles available for simulation, whatever when people start going home at 5 PM.
How powerful standard desktop computers are. There is only two orders of magnitude between a normal desktop computer (I refuse to call a Pentium III 733 as outdated) and a mainframe computer.
Now all we need are ways of getting local connections significantly faster (Did someone say Gig Ethernet) to allow faster communication between the nodes and we will be able to scale beyond several hundred and break the top 100. I hear 1gig NICs will be falling in price to under $100 US retail soon...
Why file a class action lawsuit where your lawyer will get rich and you will get $1.24 for your efforts...
You are much better off sueing either in Small Claims court,where the limits tend to be around 1-1.5K dollars, don't require a lawyer on your part, and tend to be settled pretty quickly.
Threats of filling a class action lawsuit are a waste of time, you are much better off going to your county courthouse, filling the paper work, doing a quick web search on where to send the papers, and hire a courier to deliver them to IBMs local legal representative. When the day is done they will pay you your 1500 because it is much easier/cheaper to do that then to send two lawyers at $200/hr to your location to fight it (and still loose quite a bit of money)
I have a lot of fun at work watching people attempt to take credit for each others work, teams with brilliant programmers attracing "Project Managers" like flys to sh*t, teams that can't perform loosing management resources. Why is that, it is much easier as "useless overhead" to attach yourself to a known winner and coast rather than fix a broken team and get it functional...
The problem is that as technical people we need to remove quite a lot of the need for personal glorification and let our skills shine through the projects that we participate in. When people start noticing that every project that I work on suceeds, people stop thinking it is luck, and maybe something with the teams I belong to, then decide it is somehow with the people on the teams I belong too...
While this type of respect takes a lot longer in our instant gratification society, it is much easier to keep in place once you have achieved it... Assume your professors are smart, they know from the people asking the hard questions who is doing the work, and who isn't.
Why should people avoid little hubs around their network ???
Seems cost effective to me, run one drop to every work location (rather than the 3 you suggest... everyone gets a phone) and save on cabling overall. For the small minority of people that need multiple drops, install a $40 switch on their desktop, this will save you
1) Managed switch port
2) Extra cabling costs
3) Needed space for extra wire running above/below every floor
Disadvantage - User will have to be responsible for "in cubicle" wiring... Not a bad trade-off since the people needing this will be the more tech savy people to start with
Slashdot Mirror
Now go back and look at what a Sun 10000 was doing in about 1990. If I recall it had dual 128 bit busses, each running at 50 Mhz. I am not talking about your punny desktop, small server boxes, I am talking about your large 60+ CPU server systems.
These things cost money so they don't make it into comodity hardware, but when you need the system throughput Sun is actually a pretty good company to go with.
Now just add wireless, and the ability to do H.323 and you can call me Dick T.
Seriously, this is cool. Just bought my 2 year old a $20 dollar digital camera to let her play with, not a mega pixel camera either, but good enough for her to play with and it is cheaper than film.
Isn't technology fun
I love hearing that people are finally starting to publish intentions. I have been hearing rumors about this for a year or so now, since an EVP where I worked started talking about plugging a Fibre into the side of the microprocessor (and he wanted to own that connection)
As is normal, he missed completely thinking it would be a 10GbE fiber for networking, rather than a 40+GB connection to main memory...
The comments on working on the I/O side of the processor were right on (I read the EETimes article, rather than the Register article to get "real" facts ). For years Sun was known for having the slowest RISC processor in the business, however they had the fastest boxes. No one seemed to understand this, until they realized that they were running multiple 128 bit memory buses at rather good clock rates. That was better than 10 years ago, and just now we are starting to see memory busses approaching this level in their competitors hardware.
So they are saying that if you take out a large telephone networks Central Office, people connected to this office will suffer lost connections. Infact some long distance connectivity will suffer as well.
Why does this suprise anyone. Hmmm let me see, if you take out your ISP, all of the sudden you will loose connectivity to the internet unless you pay A LOT of money to have a second line put in. Even then the chance that both of those lines run through some common area is pretty high.
Things are easy to engineer with fully redundancy, what isn't easy is to do it cheaply enough that people will still be willing to pay for it.
Of course I send them to my representatives. In the state of Oregon, I am stuck with Sen. Wyden (D- Ore) and Sen. Smith (R- Ore)and am represented by Rep David Wu (D- Ore) in the house. I write one of them at least every month on topics ranging from the environment, SPAM, technology issues, and many others.
Every time I have gotten a response from one of them. Why would you expect a response from a representative of another area anyway ? It is not their job to represent your views in congress, but those of their constituents.
Well it looks like we will never be able to download music on the internet. First it was because we weren't paying the artists, just shipping the music to each other, now it is because we have exclusive contracts with the artists to only use the one service that pays them...
I Want my Net-TV (to paraphrase Sting from a Dire Straights song)
I'd like to know where you can get a T-1 for $70 a month, if ya do, tell me where. All the Mobilstar Starbucks locations had T-1s rather than DSL or something cheaper
Sounds like a business problem to me. No wonder they were loosing money. As I said, I think you would have to sell this service to a business for under a few hundred dollars a month. Do you think starbucks cares what their ping times are, DSL is fine...
Obviously we are not paying "costs" if people are going bankrupt providing the service. I can understand if it were one maybe two badly run companies, however the whole industry is quickly going bankrupt...
You can't tell me that a $200 dollar access point, and a $70/month Broadband connection to the internet is a barrier to entry... That is all it takes to connect a starbucks to the internet. I'll start a business now, except not enough people would be willing to pay me the $200-$300/month it would cost to make this profitable enought to be interesting. (I need to make a few tens of thousand a month to pay my staff to run it, plus clear a profit for myself)
Path to Profitability.
2 years ago, it was get a customer base, then figure out how to exploit them to make a profit. Now people have realized that there are no barriers to entry, so you can't raise prices later. You have to show up front how you can make money off of each and every customer from day one, then hope you get enough of them to overcome useless overhead in the corporation (read, the CEO, CTO, C??, and much of the marketing dept.)
The days of giving away dollar bills for 3 quarters to generate revenue are over for the internet, show me how you can make money, or go the way of the Dodo bird
This is what most large companies do, buy a bunch of leased lines and run an ATM backbone (or other technology) over it to provide an internal WAN capability...
Why not do it over the internet in general ??? Well this way you have guaranteed bandwidth characteristics, as much data hiding as you want, and each office does not have to expose itself to the internet in general
One thing Microsoft does is to make sure that drivers, etc. are generally backwards compatible, sometimes to it's detriment.
I can tell you haven't written drivers for Microsoft... There are VERY large differences between 95, 95 OSR2.1 (that was the last version shipped) 98, SE, ME, and the various NTs...
The big differences are between SE/ME 9X/NT4, however there are subtle differences between NT4 and W2K, and even some fun differences between the various Service Packs for NT4. Don't even get me started with NT versions prior to 4.0.
Not a direct interview but here is an interesting article...
/. crowd can come up with
Glimpses of a Guy You'd Like to Know
Of course this took a little bit of effort to read the linux kernel mail list. Interesting start the first paragraph is
The chances are that you've seen an interview or two, in print or on television, with the fellow who a decade ago had the sheer audacity to sit down and start writing his own version of Unix. If you have, you've probably been impressed with a polite and self-assured young man who was probably giving rote answers to the same questions that are asked in every interview.
Hmmm... seems to backup my point, maybe it is a time for a slash interview and see what the
Actually more humourously, hack the phone to return a GPS coordinate of you standing RIGHT BEHIND THEM...
Oh wait, we aren't watching Real Genius are we, never mind
Actually I did this in the mid 80s to keep my aging PC working...
Install memory extender to allow 1 MB RAM
Boot off of floppy A (Did I say REALLY old PC - Before hard drives even)
Install EMS memory handler to get at memory
Install RAM DISK of size 512 KB
Copy contents of floppy B onto RAM disk (basically Borland Turbo C 1.x)
Replace floppy B with modem software
Call campus network and connect to VAX systems
Leave computer on in dorm room for months at a time.
Basically this gave me VERY fast access to a compiler (remember when you could fit Turbo C onto a single floppy), and room for temp storage as well...
When a journalist can't come up with an origional question, they seem to want to come up with a question that they know might generate controversy if it is answered, or not answered.
Do you have a beef with RMS over GNU/Linux ?
Do we have ground troops in Afghanistan ?
Have you had sexual relations with an Intern ?
When will journalists learn to at least ask a good question.
What do you think RMSs biggest contribution to the Open source movement is ?
Who inspires you today, who do you see as your idol ?
What message would you like to deliver to todays incoming college Computer Science Freshmen, what do you think they should be looking at ?
What are your choices... I mean MS upgrades their OS on a yearly bassis (to almost no effect) first it was Win 95 (and a OEM only service release every year after that) then Win 98, Win 98 SE, Win ME, Win XP... Seems pretty frequent to me...
As for not being able to upgrade because your binaries only work with a given version... Well that is why you either get source, or go with a company that will keep its software updated (and probabaly pay maintenence on the software as well)
I was glad to see that development on the 2.5 kernel will begin within the month. I know it is VERY important to get the stable kernels as stable as possible before you release the hordes on the next project. I think Linus deserves praise for keeping the 2.5 kernel back until he has as close to perfection as he can get on the 2.4 kernel.
Once he is done there, it is time to start adding features again... Too bad microsoft only sends out a new service pack about every 6-12 months vs. the 10 kernel updates that have happened since the 2.4 kernel came out last year, maybe if they could do that they might be able to put an IIS server on the internet without it being hacked from here to the far reaches of space.
Maybe it is time to download 2.4.11 and get my kernel updated from 2.4.4
Well here I thought we were staying with commodity hardware, but if we are going to go beyond commodity hardware into stuff that is engineered to provide low latency, high bandwidth, lets look at technologies like Infiniband which is engineered to avoid almost all OS latencies by delivering data directly to applications from hardware with OS bypass (almost no software between your app and the hardware anymore), and at 2 Gb/sec it isn't too bad on the performance scale. However I am betting that when this stuff comes out it will be a little pricey and go slightly over the 210,000 dollar budget
That is a serious problem though, and one I assume Beowulf clusters will take care of, what if a node goes down in the middle of processing, how does the cluster respond to it ?
The ultimate Linux selling tool, every linux box in your company is a node in a cluster, add a few servers for extra speed, add a few computers to provide file I/O and backup capability, and you have one of the fastest supercomputers available to your company without having to spend an extra dime (everyone needs a desktop anyway). Can you imagine the extra cycles available for simulation, whatever when people start going home at 5 PM.
Now all we need are ways of getting local connections significantly faster (Did someone say Gig Ethernet) to allow faster communication between the nodes and we will be able to scale beyond several hundred and break the top 100. I hear 1gig NICs will be falling in price to under $100 US retail soon...
How fast do you connect to your cluster ?
You are much better off sueing either in Small Claims court,where the limits tend to be around 1-1.5K dollars, don't require a lawyer on your part, and tend to be settled pretty quickly.
Threats of filling a class action lawsuit are a waste of time, you are much better off going to your county courthouse, filling the paper work, doing a quick web search on where to send the papers, and hire a courier to deliver them to IBMs local legal representative. When the day is done they will pay you your 1500 because it is much easier/cheaper to do that then to send two lawyers at $200/hr to your location to fight it (and still loose quite a bit of money)
I have a lot of fun at work watching people attempt to take credit for each others work, teams with brilliant programmers attracing "Project Managers" like flys to sh*t, teams that can't perform loosing management resources. Why is that, it is much easier as "useless overhead" to attach yourself to a known winner and coast rather than fix a broken team and get it functional...
The problem is that as technical people we need to remove quite a lot of the need for personal glorification and let our skills shine through the projects that we participate in. When people start noticing that every project that I work on suceeds, people stop thinking it is luck, and maybe something with the teams I belong to, then decide it is somehow with the people on the teams I belong too...
While this type of respect takes a lot longer in our instant gratification society, it is much easier to keep in place once you have achieved it... Assume your professors are smart, they know from the people asking the hard questions who is doing the work, and who isn't.
Why should people avoid little hubs around their network ???
Seems cost effective to me, run one drop to every work location (rather than the 3 you suggest... everyone gets a phone) and save on cabling overall. For the small minority of people that need multiple drops, install a $40 switch on their desktop, this will save you
1) Managed switch port
2) Extra cabling costs
3) Needed space for extra wire running above/below every floor
Disadvantage - User will have to be responsible for "in cubicle" wiring... Not a bad trade-off since the people needing this will be the more tech savy people to start with
No such luck. I wonder if we