I tend to keep up with the latest ACM/IEEE published papers. But in 20+ years of IT programming I've only met a handful of others that do anything similar.
Basically I read an interesting paper, suggest we use some small part of it at work, everybody laughs. Two years later I'm called in to explain what the hell I was talking about as it's now seen as the thing to do. (sigh)
For instance, I recently commented that augmented reality is ready to make a move into the corporate world. Laughter all round.
The very same day I was asked to re-explain the benefits of open source as the CEO had been told it would reduce the cost of IT. I just gave the same chat I did three years ago:-)
OK,
the price is to much currently and I'm not sure it has mass market appeal. But for the likes of me it would be a great idea. Especially if it had a mips or powerpc module.
Currently I have a noisy half rack of near identical machines except for the cpu:-(
Coming from many years in programming I've recently been learning about hardware. Being of a lazy disposition I needed to quickly and cheaply build up a large stock of components. Nothing I hate more than not having a part on hand.
I picked up a breadboard locally and that was the last item I bought retail. Everything else came off ebay at ridiculous prices. The problem for me was usually I ended up with too many items.
The best thing about the cost is that I'm not afraid to break things. That burning smell doesn't mean I've blown $10 more like 10c.
For robotics you can go with the latest pic/stamp + prgrammer. But think simple, Z80 + EEPROM + SRAM gives you the same sort of thing at a much reduced price. For instance I picked up 100+ Z80 chips for less than $10.
Decide which controller you're going to use, I'd suggest Z80/6502 etc for cost reasons, but the low end pics are very reasonable as well. Once you have that decided, there are numerous emulators etc available for Linux for each controller. Cheap serial burner can be built for pennies, make that the first class. Kids hate theory.
Build the course around the controller and only delve into theory briefly. Show how to build a drive controller, light/ir/sound detector etc etc. Make one class cover a topic. Give weblinks so the interested kids can follow up in there own time.
We still use Notes in a worldwide corporation that has over 25,000 seats.
We use ZOS, OS/400 & AIX and have done so for the last 10+ years, with no downtime caused by OS failure. In fact I can only remember one outgage caused by these servers, somebody ignored some disk pack erorrs when they should have called the engineer for a replacement. Needless to say said person was lucky to keep their job and was on probabtion for 6 months.
We use DB/2 and have never ever lost a record or had any downtime caused by DBMS failure.
We use MQSeries and have never ever lost a transaction or had down time caused by messaging software fail.
So prehaps you might wish to think a little larger when looking at IBM software. In fact it's hard to think of another company which provides such high quality enterprise software. The downside is cost measured in many many $$$$.
'Compliance will ultimately cost nothing...' Completely disagree with this drivel. It's costing us time and money now, it will cost us time and money in the future. CEOs/CFOs already have enough information in front of them to make decisions. If they don't they're removed and the next CEO/CFO makes his first task to have this information. Every succesful company knows their figures.
'Don't have any...' So we're spending all this money with no idea if it's going to bring a net return? Ludicrous. What if it's costing more than all the fraud etc. was before? I also seriously doubt you are able to prove that financial fraud costs less or more than terrorism to the US on an annual basis.
Alan Kay should revise his stastement that the best way to predict the future is to invent it, maybe to, the best way to predict the future is to wait until it happens.
How much does compliance cost? If you are basing your views on cost analysis you cannot omit one of the major data points?
Where is your proof that this new law will prevent future accounting mis-practice? where is your facts that this was the least costly way to go about it.
The truth of the matter is that this legislation was knee jerk reaction to a complex and deep issue. For instance compliance only applies to publicly listed companies. The legislation does not take into account how other countries audit companies. etc etc ad infinitum.
Lastly, whenever paperwork is introduced it never goes away. The correct answer was for _ALL_ companies trading in the US to publish complete annual accounts down to the transaction level. This would be easier for the company, would provide all the detail any investigator needed.
I didn't go to college but ended up with a degree anyway. I was advised to attend college by just about everybody, but I started work and took my degree postal with an accredited University.
If you can get accepted to a university do so and then delay entry for a year, most colleges will allow this.
During the year off either travel or work like a dog. Either way you'll be in a better psotion to judge next year if you want to go.
And you've yet to grasp the obvious significance of this. Let's see. We send information to aliens and get no response, we send information to IT guys and get no response, therefore IT guys=Aliens.
>> Interesting, I run Debian, with kde 3.4 over >> kernel 2.6.10 and my distro does not feel 5 years >> out of date.
I would guess you're not running stable or testing but unstable. I run testing and it's too far behind the idea of release early and often. I'll probably go to unstable this evening.
Debian takes too long to do releases. It's not NetBSD it should change to a tiered release structure. The four mentioned are a good idea.
In short the time frame between Debian releases is indefensible, it takes to long.
Given how quickly you judge others and applaud their mis-fortune I can only hope that your skeletons jump out of their respective cupboards and beat some sense in to you.
Ethics in business. What next truthfulness in politics? Fairness in the legal profession? Forgiveness in religion?
My guess is Harvard is going to have some pretty large legal bills to pay:-)
Slashdot Mirror
I tend to keep up with the latest ACM/IEEE published papers. But in 20+ years of IT programming I've only met a handful of others that do anything similar.
:-)
Basically I read an interesting paper, suggest we use some small part of it at work, everybody laughs. Two years later I'm called in to explain what the hell I was talking about as it's now seen as the thing to do. (sigh)
For instance, I recently commented that augmented reality is ready to make a move into the corporate world. Laughter all round.
The very same day I was asked to re-explain the benefits of open source as the CEO had been told it would reduce the cost of IT. I just gave the same chat I did three years ago
Send them the file and they'll return a bound manual.
Will somebody just fire the mods or better still start a /. MKII
This would appear to mesh exactly with your aims :-
http://www.opencroquet.org/
Two psossible downsides SmallTalk and Alan Kay. Both are positives for me but I understand some peoples aversion.
Don't tell em yer name Spike
Those three would rip each other to shreds before they realised there was anyone else in the room!
OK,
:-(
the price is to much currently and I'm not sure it has mass market appeal. But for the likes of me it would be a great idea. Especially if it had a mips or powerpc module.
Currently I have a noisy half rack of near identical machines except for the cpu
Bonus points for telling us where he pinched the idea from?
Coming from many years in programming I've recently been learning about hardware. Being of a lazy disposition I needed to quickly and cheaply build up a large stock of components. Nothing I hate more than not having a part on hand.
I picked up a breadboard locally and that was the last item I bought retail. Everything else came off ebay at ridiculous prices. The problem for me was usually I ended up with too many items.
The best thing about the cost is that I'm not afraid to break things. That burning smell doesn't mean I've blown $10 more like 10c.
For robotics you can go with the latest pic/stamp + prgrammer. But think simple, Z80 + EEPROM + SRAM gives you the same sort of thing at a much reduced price. For instance I picked up 100+ Z80 chips for less than $10.
Decide which controller you're going to use, I'd suggest Z80/6502 etc for cost reasons, but the low end pics are very reasonable as well. Once you have that decided, there are numerous emulators etc available for Linux for each controller. Cheap serial burner can be built for pennies, make that the first class. Kids hate theory.
Build the course around the controller and only delve into theory briefly. Show how to build a drive controller, light/ir/sound detector etc etc. Make one class cover a topic. Give weblinks so the interested kids can follow up in there own time.
I mean they have the chance to sell a reasonably secure OS and insted they do this. Hanging's to good for em.
We still use Notes in a worldwide corporation that has over 25,000 seats.
We use ZOS, OS/400 & AIX and have done so for the last 10+ years, with no downtime caused by OS failure. In fact I can only remember one outgage caused by these servers, somebody ignored some disk pack erorrs when they should have called the engineer for a replacement. Needless to say said person was lucky to keep their job and was on probabtion for 6 months.
We use DB/2 and have never ever lost a record or had any downtime caused by DBMS failure.
We use MQSeries and have never ever lost a transaction or had down time caused by messaging software fail.
So prehaps you might wish to think a little larger when looking at IBM software. In fact it's hard to think of another company which provides such high quality enterprise software. The downside is cost measured in many many $$$$.
Hello SmallTalk, CLOS and Ruby
'Compliance will ultimately cost nothing...' Completely disagree with this drivel. It's costing us time and money now, it will cost us time and money in the future. CEOs/CFOs already have enough information in front of them to make decisions. If they don't they're removed and the next CEO/CFO makes his first task to have this information. Every succesful company knows their figures.
'Don't have any...' So we're spending all this money with no idea if it's going to bring a net return? Ludicrous. What if it's costing more than all the fraud etc. was before? I also seriously doubt you are able to prove that financial fraud costs less or more than terrorism to the US on an annual basis.
Alan Kay should revise his stastement that the best way to predict the future is to invent it, maybe to, the best way to predict the future is to wait until it happens.
How much does compliance cost? If you are basing your views on cost analysis you cannot omit one of the major data points?
Where is your proof that this new law will prevent future accounting mis-practice? where is your facts that this was the least costly way to go about it.
The truth of the matter is that this legislation was knee jerk reaction to a complex and deep issue. For instance compliance only applies to publicly listed companies. The legislation does not take into account how other countries audit companies. etc etc ad infinitum.
Lastly, whenever paperwork is introduced it never goes away. The correct answer was for _ALL_ companies trading in the US to publish complete annual accounts down to the transaction level. This would be easier for the company, would provide all the detail any investigator needed.
When I talked about OO, C++ was not what I had in mind.
Thank god the kernel is not written in C.
I didn't go to college but ended up with a degree anyway. I was advised to attend college by just about everybody, but I started work and took my degree postal with an accredited University.
If you can get accepted to a university do so and then delay entry for a year, most colleges will allow this.
During the year off either travel or work like a dog. Either way you'll be in a better psotion to judge next year if you want to go.
Where can I buy a notebook with > 8 hr battery life?
I'd give up the CD/DVD, the color screen, the ghz proc. I'd give up most things to get a decent battery life. Now the ideal would be about 40hrs.
Any ideas?
And you've yet to grasp the obvious significance of this. Let's see. We send information to aliens and get no response, we send information to IT guys and get no response, therefore IT guys=Aliens.
I suggest you look up fidelity in a dictionary. Fidelity ~ Quality.
Take a look at the mesh networking stuff. Do not think about having one central access point. Every home should have it's own node on the mesh.
I've said it before and I'll say it again. Timothy is a useless cretin, I could write a bot that does a better job than him. Fire his useless ass.
A permanent vacation would be better for all.
>> Interesting, I run Debian, with kde 3.4 over
>> kernel 2.6.10 and my distro does not feel 5 years
>> out of date.
I would guess you're not running stable or testing but unstable. I run testing and it's too far behind the idea of release early and often. I'll probably go to unstable this evening.
Debian takes too long to do releases. It's not NetBSD it should change to a tiered release structure. The four mentioned are a good idea.
In short the time frame between Debian releases is indefensible, it takes to long.
Given how quickly you judge others and applaud their mis-fortune I can only hope that your skeletons jump out of their respective cupboards and beat some sense in to you.
:-)
Ethics in business. What next truthfulness in politics? Fairness in the legal profession? Forgiveness in religion?
My guess is Harvard is going to have some pretty large legal bills to pay