A long time ago (in 'Net time anyway), there was a Slashdot article regarding what IBM could do to help the open-source world, now that they are heavily into Linux. Here's one place they may be able to help, through experience if not actual code.
On z/OS operating systems, there are two (or three) logging architectures: there are instructions and supervisor calls for low-level tracing (Generalized Trace Facility) which is a high-overhead debugging tool. There's the System Management Facility, which is a software standard for logging things like utilization data, events such as file open/close (with I/O or media error statistics), etcetera. SMF would be the closest to 'normal' logging.
There's then a 'System Logger' facility which handles log files of all sorts, cycling them, etc.
This kind of standardized logging architecture might be a good thing for Linux, also.
I work for a state government, we "outsource" stuff to subcontractors sometimes, too. Here's the problem: subcontracting removes the ability to hold someone accountable. The manager (it's always a manager, they fired the employees when they outsourced it) responsible for the program says "don't look at me! It's the subcontractor's fault". Yet the government is usually powerless to tell the subcontractor to fire the idiots responsible ("government interference with 'small' business! Oh my god!"). A third problem: the subcontractor makes their profit by doing things cheaply, so they pay the employees as little as possible, creating a bribery incentive for data like this to be copied.
I don't really like the TSA, but if we _have_ to have it, none of the operations should be outsourced. They should all be direct government employees without career service protections, so they can be fired at will. Perhaps put them into a branch of the military (Coast Guard would be the closest in terms of purpose), and if they screw up, rotate them into a combat zone somewhere - or, if we aren't in conflicts anywhere, let them guard radar installations at Point Barrow, Alaska. Whatever it takes to make the employees know that their performance is judged _seriously_.
The better solution is to abolish it and therefore the cost of running it... but like I said, if we have to have it, then let's at least get serious about doing it right.
Gorillaz with a 'z'. Grammy Awards, 2005 I think. The way it was staged, I'm not sure where the 'foil' was, since Madonna walked in front of one animated character and behind another (or perhaps I was looking at her legs or something and was fooled).
As others have stated, not holographic but still interesting.
Musion's site has videos of several interesting examples.
This won't apply if you're living alone, but in family or cooperative-living arrangements there are usually two (or more) vehicles. That means that you can easily buy one limited range vehicle for commuting, while retaining the second vehicle which would be used for long range driving, and if necessary, less-efficient commuting trips.
Let me also recommend to "you young whippersnappers" that you check out careers in the mainframe world.
The mainframe's far from dead, some young people have noticed, and as we older people retire from our system administration posts, there will be a demand for people with the foresight to acquire the skills. Where there's a demand for skill sets and a shortage of people with those skills, salaries go up.
At one time there was an extremely long term storage solution called (I think) HD-ROM, which was discs of nickel etched by an ion beam. An implementation seems to be at this link http://ww.norsam.com/hdrosetta.htm/
How is this different from an EEPROM memory cell, technology which has been around for a long time? I would submit that we already have the ability to retain memory when the power is off!
landline plans start at around 15 USD a month, cell phone plans start at what, 39 USD? So, total dollars spent is meaningless except as a metric for potential businesses to see how much money they can make.
It's similar to comparing box office dollar amounts between years -- if the ticket prices is higher in one year than the other, then total dollar comparisons don't reveal anything about usage.
A better metric would be the number of land line accounts vs. cell phone accounts
I am a systems programmer for a mainframe system which is basically the same thing as a sysadmin for *nix or Windows systems. We have been asked before about such metrics (although the PHB seems to have been distracted by other things for a while).
I haven't read all of the replies, so this may have been covered, but when we were thinking about this, all of our best metrics were "negative" ones with zero being the ideal number for the metric:
Number of times the system failed during 'prime time' Number of times users were unable to connect to the system during published hours of availability Number of times backups failed to run Number of restore requests which failed due to a bad backup Number of intervals where response time for application 'X' exceeded the 'good' value
Etcetera. Our jobs as administrators of systems used by others is to ensure that bad things do not happen, therefore it seems logical to measure or count the bad things, and report on those. When we're doing our jobs well, no one should notice us.
In which he describes using current TV methods to distribute content for "computers" (I put it in quotes because this would include Tivos and the like which to consumers are different but to cognescenti are nothing more than computers with a TV-viewing-focused UI). Between Cuban and Gates, what I think we're hearing is that television will morph from its current broadcaster/cable-company- controlled state to a state where viewers of video content control when and where they view it, and what devices they use to view it. Most of the technological challenges are already solved, the main issues remaining to be resolved revolve around financial aspects, and they're not simple: if you control the viewing, you also control the ability to skip advertising. If there's no advertisng, who pays for the content? Subscription models probably won't work, at least not currently, because there is enough free content out there to fill someone's available viewing time (and don't start on the quality argument; free content on the 'Net is demonstrably superior to say, 'Dancing with the Stars!)
The System 360/370 series, and their successors, solved the "memory overflow" problem decades ago by a combination of hardware and OS architecture. Each page of real storage in hardware has a "storage bump" containing, along with the recently-used bit that enables paging, a storage key. Programs do not allocate memory - instead they make OS calls and the OS allocates memory. It allocates this memory in the key of the program being executed, usually. OS and system-level programs execute in keys 0-7, and non-OS programs execute in key 8. Things like communication buffers and the like are allocated by system functions, not by your application program; and if your key 8 program tries to store into a key 0-7 area without calling some authorized system function to do so, your program is terminated. This mechanism prevents a lot of malicious stuff from happening because you cannot generally store into any OS-related areas. You can, of course, modify your own program via array overflows or whatever, but depending upon various ways you bound / linkedited your program that also could get it killed.
I doubt that the current set of applications in Windows world will allow migration to the above kind of architecture on Intel hardware - but the example is there for any who want to follow it.
I wonder if this guy is upset because his particular subset of Yahoo! didn't get all of the budget that he thought they should?
I hear the people that say Yahoo! should focus on something "great" - but you know what, a lot of people who say that really mean "do something that makes the stock price go up, so I can unload and make a killing rather than the profit I would currently make, which I feel is meager".
Yahoo does a lot of things that are "good enough". Yes, they could use improvements, but on the other hand most of them "just work" and their target market is not the Slashdot crowd of technological bleeding-edge early adopters, but closer to Grandpa and Grandma who want to use "that Internet thing".. and for them, Yahoo works pretty well (so did AOL for that matter I guess).
They also do some other things that may be of more long-term benefit to everyone, things like being a member of the Calendaring & Scheduling Consortium, a group trying to help solve the calendar interoperability problems which still plague us all. That might not be of immediate benefit, but sometime in the relatively near future you'll be able to schedule time with someone no matter what calendar tool they user - and that will be a major deal.
I don't think there's a lot of customer loyalty to be gained by anyone in the WiFi business. People use WiFi like they use a water fountain or the lights in a business, although for longer periods of time. They will come to a coffee shop, buy a cup of coffee, and hang out for quite a while using the AP. But it doesn't make them loyal to that coffee shop any more than using the restroom in there does - and a restroom is a fairly decent analogy because most people use a WiFi connection when they feel the urge and wherever they find a facility; they don't wait to go until they get to their "favorite" location.
Apdex is an interesting way of scoring application time, which assigns response times a score from 0 to 1 depending upon whether they meet their "tolerance" goals or not. For any transaction, there's an number T which is the upper limit that users will tolerate for response time, and a number F, where F=4T, that is the number at which people consider the system to be failing. In the study mentioned it sounds like T=1(calculation of F is left as an excercise for the reader). Once you have these numbers, you collect statistics on the response in three buckets: total transactions, satisfied count where response is T and F. The Apdex number is (satisfied count plus half of the tolerating count) divided by the total responses.
The advantage to this system is that it gives you a number telling you how well something is working in relation to its goals, and it works no matter what the units of time used are.
I live in Tallahassee, which despite being the center of election controversy in 2000, has a very good _local_ elections supervisor - in fact, he had Black Box Voting in to test some electronic machines despite it stirring controversy with his bosses. In our local elections for years we've used the optical scan system, where you darken in ovals and then feed the paper to a scanner, and we've had a very low rate of errors and problems in our elections.
My question is - what are the risks and security holes related to optical sense systems? They do provide a paper trail, but after the scanning I imagine there must be some risks with the data after that.
Do you offer the opportunity for someone to work for you as a developer while living outside the high-tech corridors which appear to be very expensive, cost-of-living-wise? Could someone do development for you while living in Montana, for example?
Didn't see this mentioned (could have missed it)-
on
Oracle Linux?
·
· Score: 2, Interesting
This seems like an move for Oracle to be able to offer a "database appliance": prebuild boxes with Linux and Oracle, and sell those to PHBs as "drop in" solutions. Many in management would fall for it.
it will learn to improvise off others melodies, leading to it being classified as one of those "crazy" jazz musicians and anyone who's ever lifted a soldering gun will be vilified in a movie called "Resistor Madness"
When asked to produce a CD, burn one from the relevant data (or.ISO image of the original)on the server. Copy originals to server as they arrive and then destroy (for privacy/security concerns)
If some auditor has a problem, then say "our reading of the policy (law, rule, whatever) was that the _data_ had to be retained, not the actual medium it arrived on".
Doesn't solve the OP's problem, but defers dealing with thousands of CD's until the first time it's a major audit problem.
"The problem lies in that the Universe cannot have both MECOs and black holes - it can only have one or the other. If this object truly is a MECO, then black holes do not exist."
OR - our current theories and understandings are wrong.
Slashdot Mirror
A long time ago (in 'Net time anyway), there was a Slashdot article regarding what IBM could do to help the open-source world, now that they are heavily into Linux. Here's one place they may be able to help, through experience if not actual code. On z/OS operating systems, there are two (or three) logging architectures: there are instructions and supervisor calls for low-level tracing (Generalized Trace Facility) which is a high-overhead debugging tool. There's the System Management Facility, which is a software standard for logging things like utilization data, events such as file open/close (with I/O or media error statistics), etcetera. SMF would be the closest to 'normal' logging. There's then a 'System Logger' facility which handles log files of all sorts, cycling them, etc. This kind of standardized logging architecture might be a good thing for Linux, also.
I work for a state government, we "outsource" stuff to subcontractors sometimes, too. Here's the problem: subcontracting removes the ability to hold someone accountable. The manager (it's always a manager, they fired the employees when they outsourced it) responsible for the program says "don't look at me! It's the subcontractor's fault". Yet the government is usually powerless to tell the subcontractor to fire the idiots responsible ("government interference with 'small' business! Oh my god!"). A third problem: the subcontractor makes their profit by doing things cheaply, so they pay the employees as little as possible, creating a bribery incentive for data like this to be copied. I don't really like the TSA, but if we _have_ to have it, none of the operations should be outsourced. They should all be direct government employees without career service protections, so they can be fired at will. Perhaps put them into a branch of the military (Coast Guard would be the closest in terms of purpose), and if they screw up, rotate them into a combat zone somewhere - or, if we aren't in conflicts anywhere, let them guard radar installations at Point Barrow, Alaska. Whatever it takes to make the employees know that their performance is judged _seriously_. The better solution is to abolish it and therefore the cost of running it... but like I said, if we have to have it, then let's at least get serious about doing it right.
It's the TSA (well really their subcontractor)! They were busy looking at the new body-scanner images of the cute folks who'd been through that week.
Aren't we all distendants of the homonyms who fell out of the trees long ago, and learned to walk up write?
Gorillaz with a 'z'. Grammy Awards, 2005 I think. The way it was staged, I'm not sure where the 'foil' was, since Madonna walked in front of one animated character and behind another (or perhaps I was looking at her legs or something and was fooled). As others have stated, not holographic but still interesting. Musion's site has videos of several interesting examples.
This won't apply if you're living alone, but in family or cooperative-living arrangements there are usually two (or more) vehicles. That means that you can easily buy one limited range vehicle for commuting, while retaining the second vehicle which would be used for long range driving, and if necessary, less-efficient commuting trips.
Let me also recommend to "you young whippersnappers" that you check out careers in the mainframe world. The mainframe's far from dead, some young people have noticed, and as we older people retire from our system administration posts, there will be a demand for people with the foresight to acquire the skills. Where there's a demand for skill sets and a shortage of people with those skills, salaries go up.
At one time there was an extremely long term storage solution called (I think) HD-ROM, which was discs of nickel etched by an ion beam. An implementation seems to be at this link http://ww.norsam.com/hdrosetta.htm/
Breathe In. Breathe Out. Eat. Eliminate. And we still don't have the bugs out!
How is this different from an EEPROM memory cell, technology which has been around for a long time? I would submit that we already have the ability to retain memory when the power is off!
Question from someone not totally up on cell-phone technology: do SMS messages include location information when sent ?
landline plans start at around 15 USD a month, cell phone plans start at what, 39 USD? So, total dollars spent is meaningless except as a metric for potential businesses to see how much money they can make. It's similar to comparing box office dollar amounts between years -- if the ticket prices is higher in one year than the other, then total dollar comparisons don't reveal anything about usage. A better metric would be the number of land line accounts vs. cell phone accounts
I am a systems programmer for a mainframe system which is basically the same thing as a sysadmin for *nix or Windows systems. We have been asked before about such metrics (although the PHB seems to have been distracted by other things for a while).
I haven't read all of the replies, so this may have been covered, but when we were thinking about this, all of our best metrics were "negative" ones with zero being the ideal number for the metric:
Number of times the system failed during 'prime time'
Number of times users were unable to connect to the system during published hours of availability
Number of times backups failed to run
Number of restore requests which failed due to a bad backup
Number of intervals where response time for application 'X' exceeded the 'good' value
Etcetera. Our jobs as administrators of systems used by others is to ensure that bad things do not happen, therefore it seems logical to measure or count the bad things, and report on those. When we're doing our jobs well, no one should notice us.
In which he describes using current TV methods to distribute content for "computers" (I put it in quotes because this would include Tivos and the like which to consumers are different but to cognescenti are nothing more than computers with a TV-viewing-focused UI). Between Cuban and Gates, what I think we're hearing is that television will morph from its current broadcaster/cable-company- controlled state to a state where viewers of video content control when and where they view it, and what devices they use to view it. Most of the technological challenges are already solved, the main issues remaining to be resolved revolve around financial aspects, and they're not simple: if you control the viewing, you also control the ability to skip advertising. If there's no advertisng, who pays for the content? Subscription models probably won't work, at least not currently, because there is enough free content out there to fill someone's available viewing time (and don't start on the quality argument; free content on the 'Net is demonstrably superior to say, 'Dancing with the Stars!)
The System 360/370 series, and their successors, solved the "memory overflow" problem decades ago by a combination of hardware and OS architecture. Each page of real storage in hardware has a "storage bump" containing, along with the recently-used bit that enables paging, a storage key. Programs do not allocate memory - instead they make OS calls and the OS allocates memory. It allocates this memory in the key of the program being executed, usually. OS and system-level programs execute in keys 0-7, and non-OS programs execute in key 8. Things like communication buffers and the like are allocated by system functions, not by your application program; and if your key 8 program tries to store into a key 0-7 area without calling some authorized system function to do so, your program is terminated. This mechanism prevents a lot of malicious stuff from happening because you cannot generally store into any OS-related areas. You can, of course, modify your own program via array overflows or whatever, but depending upon various ways you bound / linkedited your program that also could get it killed. I doubt that the current set of applications in Windows world will allow migration to the above kind of architecture on Intel hardware - but the example is there for any who want to follow it.
I wonder if this guy is upset because his particular subset of Yahoo! didn't get all of the budget that he thought they should? I hear the people that say Yahoo! should focus on something "great" - but you know what, a lot of people who say that really mean "do something that makes the stock price go up, so I can unload and make a killing rather than the profit I would currently make, which I feel is meager". Yahoo does a lot of things that are "good enough". Yes, they could use improvements, but on the other hand most of them "just work" and their target market is not the Slashdot crowd of technological bleeding-edge early adopters, but closer to Grandpa and Grandma who want to use "that Internet thing".. and for them, Yahoo works pretty well (so did AOL for that matter I guess). They also do some other things that may be of more long-term benefit to everyone, things like being a member of the Calendaring & Scheduling Consortium, a group trying to help solve the calendar interoperability problems which still plague us all. That might not be of immediate benefit, but sometime in the relatively near future you'll be able to schedule time with someone no matter what calendar tool they user - and that will be a major deal.
I don't think there's a lot of customer loyalty to be gained by anyone in the WiFi business. People use WiFi like they use a water fountain or the lights in a business, although for longer periods of time. They will come to a coffee shop, buy a cup of coffee, and hang out for quite a while using the AP. But it doesn't make them loyal to that coffee shop any more than using the restroom in there does - and a restroom is a fairly decent analogy because most people use a WiFi connection when they feel the urge and wherever they find a facility; they don't wait to go until they get to their "favorite" location.
Apdex is an interesting way of scoring application time, which assigns response times a score from 0 to 1 depending upon whether they meet their "tolerance" goals or not. For any transaction, there's an number T which is the upper limit that users will tolerate for response time, and a number F, where F=4T, that is the number at which people consider the system to be failing. In the study mentioned it sounds like T=1(calculation of F is left as an excercise for the reader). Once you have these numbers, you collect statistics on the response in three buckets: total transactions, satisfied count where response is T and F. The Apdex number is (satisfied count plus half of the tolerating count) divided by the total responses. The advantage to this system is that it gives you a number telling you how well something is working in relation to its goals, and it works no matter what the units of time used are.
I live in Tallahassee, which despite being the center of election controversy in 2000, has a very good _local_ elections supervisor - in fact, he had Black Box Voting in to test some electronic machines despite it stirring controversy with his bosses. In our local elections for years we've used the optical scan system, where you darken in ovals and then feed the paper to a scanner, and we've had a very low rate of errors and problems in our elections. My question is - what are the risks and security holes related to optical sense systems? They do provide a paper trail, but after the scanning I imagine there must be some risks with the data after that.
Do you offer the opportunity for someone to work for you as a developer while living outside the high-tech corridors which appear to be very expensive, cost-of-living-wise? Could someone do development for you while living in Montana, for example?
This seems like an move for Oracle to be able to offer a "database appliance": prebuild boxes with Linux and Oracle, and sell those to PHBs as "drop in" solutions. Many in management would fall for it.
it will learn to improvise off others melodies, leading to it being classified as one of those "crazy" jazz musicians and anyone who's ever lifted a soldering gun will be vilified in a movie called "Resistor Madness"
"That rock looks only meters away! No, I mean yardss... AW CRAP NOT AGAIN!"
When asked to produce a CD, burn one from the relevant data (or .ISO image of the original)on the server. Copy originals to server as they arrive and then destroy (for privacy/security concerns)
If some auditor has a problem, then say "our reading of the policy (law, rule, whatever) was that the _data_ had to be retained, not the actual medium it arrived on".
Doesn't solve the OP's problem, but defers dealing with thousands of CD's until the first time it's a major audit problem.
"The problem lies in that the Universe cannot have both MECOs and black holes - it can only have one or the other. If this object truly is a MECO, then black holes do not exist." OR - our current theories and understandings are wrong.