Yes, I just mean that the IRQ interface as it's implimented is way lame.
Put more than a couple RAID/Network etc cards in your server, and poof, all IRQ's are used. Now what?
What I really mean, is to make a "IRQ" system that works right. It's kind of like having 256MB of RAM under windows 3.1, but still running out of GDI/User/System resources - even though lots or RAM is still free...
How about getting rid of IRQ's on the PC platform!
How about getting rid of drive letters in Windows/Dos and having mount points!
How about a better drive interface than the stupid IDE interface. (Macs did it right with SCSI, but now to be "cheap" they do it too [sigh])
And for self healing? If Windows is still around and the predominant OS, I'll pass on the "self healing" - it'll be more like "death-without-dignity." Remember NT 4 SP 6? [Shivver] I don't want MS "self-healing" my machine!
In fact, I don't think I want anyone self healing my machine until software is lots more robust than it is now. At least when I apply patches to my machine and notice that something isn't working right, I know I _just_ patched it, so it might be the patch. With someone else applying patches without my knowing, I would be screwed!
Yeah, all those "wonderful things are just around the corner" articles are neat, but I would truly be happy with some "incremental" changes.
Lets forget "visionary" for a while and just fix the crap that's broken right now! Pleeeeease!
Ever notice how the "tort reform" people always point out how some moron who cut his hair with the lawn-mower got 43 gazillion dollars, and thus the system needs reform?
You never hear about the harrassing lawsuits large corporations bring on each other...and in fact, in all the legislation I've heard about, the changes desired were to limit pain & suffering etc. They never limit the ability for big biz to sue and reap huge rewards against each other.
Either software is entirely different or it isn't.
Go read some good material on software design. "The Software Conspiracy" by Mark Minasi (I spell horribly, so beware)
Well designed software isn't that much more expensive. Until it gets done regularly, it will be hard to tell how much more expensive it will be, but I can see it potentially costing less.
I don't demand flawless, but then tort law doesn't either. It just demands a reasonable standard. What's reasonable, is what you can convince a jury of...
Software is continually ruling our lives...
Would it be OK, if your car just stopped regularly? It's just software, sure, it controls the fuel/air mixture, but hey it's just electrons running around anyway...
We'll get decent software when we demand that it function "just like everything else other than software." We expect more from the hardware that we run said software on. We wouldn't tolerate cars or microwaves or cell phones or PS2's that run as badly as much software does. That's just plain sad. I guess we get what we deserve.
Until the public starts to see that software design isn't black magic, that no-one can predict, and that it's artistry...blah blah blah, we won't get good software...well that's life I guess.
What is being described is just the very, very, very early, most rudimentary stages of learning about embryos, and the uterus.
One would think from the article, that "in the next few weeks, we'll be growing babies in an acquarium!"
Doctors and Scientists don't even understand some very basic problems in pregnancy in people during normal pregnancy.
Doctors have no real idea what causes HELLP/Pre-Eclampsia. See http://www.dartmouth.edu/~obgyn/mfm/PatientEd/PIH_ HELLP.html
or
http://www4.ncbi.nlm.nih.gov/PubMed/
search on HELLP
Not only that, they don't know how to stop it, or treat it with any real effectiveness. The only solution, is to hope that the baby is sufficiently developed that it will survive outside the uterus. Then they symptomatically treat the disease. (i.e. Treat blood-pressure, but can't figure out why it goes sky-high, and can't prevent this.) They keep the baby from having problems, and develop the lungs with steriods. The only solution to keep the mother and baby from dying (provided it's mature enough to live outside the uterus) is to deliver the baby.
I know this because I have several OBGYN family members, and my wife had a very serious case of HELLP. It's way serious...and science doesn't have a clue.
Thus, my skeptcism about this breakthrough. Sure, it's incredible science, and shows lots of hard work. But we're so far from actually doing what's proposed, it laughable. It's like giving a press release about your child who just learned to stand-up today, and claiming that they will be challenging Michael Johnson as a sprinter next week. Ha!
I wish them good luck, and kudo's. But just remember they have light-years left to go.
Finally, the whole process brings up a whole host of ethical issues. Lets say, they get it where they can keep embryos alive for many weeks. Do we know they are doing ok, or will they have missed some vital hormone sequence? Getting a baby alive to term is a big enough task, but will we have a normal baby?
What about the babies that reach some form of viability, but have extremely serious health problems. Do we euthanize them?
In short, we only have the faintest inkling of how things work. We haven't solved any of the ethical problems. We've got a lot of work to do before any of this could happen.
I normally end with "Cheers!" but the topic seems way too serious to be flip...
I'm sure you're right, at least in the case you describe.
The point is, that DSL is a more elegant technical solution. There are fewer problems associated, with large collision domains etc. Just like ethernet was a lousier technical solution than token-ring. Yet for other reasons, ethernet survived. (I think that's a shame...but I digress.:) )
DSL is more easily engineered to not have bandwidth oversell problems, as much as cable is. That doesn't mean that either will succeed. In fact, an on-the-ball cable co might just do it when DSL doesn't. But, just like it's easier to design a good fault tolerant, easy to maintain token-ring network, it's a whole lot easier to design DSL.
Lastly, if I don't have a say in design, which most of us don't, it's a whole lot easier to find a choice that agrees with what we need on DSL. Telco provides wires, nothing else. DSL circuit by someone else, and IP Feed from yet another. As many have pointed out, the IP feed is the most likely culprit, thus, if you can pick ISP's, it's a whole lot easier to solve the IP oversell issue. Thus DSL, is more likely to have an acceptble solution.
The BEST solution is what works...my point is that it seems more likely that DSL will be a preferred solution, especially for exoteric needs.
There was a discussion re: Smoothwall firewall, and the founder/creator, which conincendentially just happens to be Richard (or more appropriately Dick.)
From what I've seen about both guys (it might not be true, but it is from different sources, and widely accepted.) both are abrasive and rather unpleasant. (I'm not saying they're the same, but making some observations, and using both to make a point.)
I've had some friends and acquantences, who thought the whole world was wrong. And in fact, in many ways, they're right! But the way they go about "fixing" the problem - is in itself a problem. They attack everyone. They are abrasive. They are often hypocritical. They burn their bridges, and aren't polite. They use massive shock value to make a point, often offending many, including those who might help them.
Eventually, enough people are really sick of them, and practically want to burn them at the stake (or is that steak?)!
The point is this. I like pleasant people. Pleasant people don't have to compromise their principles, or sell out. But people who are not offensive and have an inflated sence of self importance are just more effective.
RMS probbaly isn't going to change, and that's too bad. From what I can see, his actions are his own worst enemy. Same with Richard from Smoothwall. They could have lots of positive effects, but they make themselves largely ineffective.
That's just too bad.
As to the hypercritical (hyper not hypo) response to them...well what's to surprise? They offend people, and then expect an evenhanded response. Sure, it would be more fair, but I know as well as anyone else, that a fair response is not very likely.
RMS by and large, helps shape the response he gets. I assume that he knows it, and doesn't care. That's the way things go.
Did you hear that he lost ancestors in the death-camps during the holocaust?
Yup, they fell out of the guard towers...
No, I don't feel sorry for JH. He's just out to grab power for the Justice Dpt. Sure, he probably thinks he's right, but so did Hitler. Thinking you're right doesn't make it so.
tort Pronunciation Key (tôrt)
n. Law
Damage, injury, or a wrongful act done willfully, negligently, or in circumstances involving strict liability, but not involving breach of contract, for which a civil suit can be brought.
This is what you sue civially for...not loss of life etc. Your definition of liability wouldn't cover my crashing into your car, as long as you didn't die, or have the real possibility of dieing. [Sheesh]
Every, let me say it again! EVERY company should face liability for the products it makes. This is an essential part of a free market system. Everyone is free to make a product, but by participating in a system where everyone is free to create suttf, the maker must ALSO agree to be responsible to a court system.
The court system is where mistakes in the creation of the product are adjudicated. If the maker/creator was negligent in the eyes of a jury, and caused damage to a party, then the creator/maker must "make it as if the damage never occured." Software makers should also suffer this liability.
Now, the hue and cry will be "Oh, software will be so expensive, if we have to eliminate most bugs - software is ALWAYS buggy!" Bah! Software _will_ cost more, if we demand a higher standard. But, that's not such a bad thing. What companies like MS (and others who make shoddy, feature laden, rather than solid software) do is shift the cost burden.
Instead of SuperDuper Wordprocessor costing $800, it only costs (up front) $250. Then the company that buys that software then spends way more fixing problems, and lost documents etc. So, the software really cost $1500 say, but mgmt thinks it only cost $250. Now mgmt should realize this, and probably they eventually will. That will change the system.
But to say that we don't need the legal system is stupid. If I produce a crap lawn-mower that wrecks your lawn, I should be liable. If I produce crap software that tells you to apply 100 Lbs of fertilizer / sqare foot, and that wrecks your lawn, I should be liable. (Free software would probably need some waiver.) What's so different there. There might be some slight modifications needed for software, but if software manufacturers were liable on a "sale of goods" basis for damage from the software, then I expect we would see some serious improvement in the quality of our software. I would be that or die.
Damn straight, and if you don't go out and CONSUME right now, I'll be forced to call Herr Ashcroft on you.
[George says]
Non consumers are EVIL, like that Al Keida guy or guys or something like that...Uhhh, where's Afgainistain again Dick? Nevermind, that stuff makes my head hurt. I'll just take another pretzel!
Yes, I didn't have the "opportunity" to use punch cards... [Groan] You could all see the "old" timer story coming couldn't you...
My prof in university told me of his first programming job. A payroll system. They didn't have a computer system yet, so they diagrammed and setup the program on punch cards. Then they took the completed program (punch-cards) and bought some spare time on another machine. After feeding the punch cards, the program ran correctly tbe first time!
Sheesh, and I ues a compiler as a syntax checker. When was the last time you got anything more complex than "hello world" to compile and run correctly the first time. (Ok, I'm a sucky programmer [grin]) But never the less, program design was a whole lot more rigerous then!
Instead of looking at the "needs" of your subordinates, first look at the company you want to work for.
Sure, finding out how to support the people under you is important, but not the most important question.
The most important question, is, "what is the company/mamangement I must work under like?"
If your company is ethical and concerned about it's people (really concerned, not just financially concerned) your job will be much easier. Then the task only becomes finding ways to help your subordinates do their jobs. You'll spend lots less time fighting management above you to actually get this priviledge. That's a huge help.
I know this sounds simplistic, but my exp in this area is that when I am empowered by the employer/upper management, I can really focus on doing what needs to be done. Lots less time is spent on CYA, political fighting, empire building etc. Then you're happy, you can be honest and upfront with your subordinates, and gain their respect and trust. (Trust, i think, is of paramount importance!) Then they'll tell you when you're doing stuff wrong, and help you from looking like a schmuck. Then you can help them get their needs met and be productive.
The end result!? The company runs smoother, more efficiently, and more profitably.
Thus, see what you're empowered to do by your managers, than when it's right, figure out what the specific needs of your subordinates are. They're never the same, but the overall principals are!
I probably won't go looking through code unless I have a reason. The #1 reason would be a bug/feature that doesn't work right. So, I do a few searches, find no solution. Look at updates etc. still no joy. I crack open the code, review it, run my paticular bug through it, and bingo I find a bug. In OSS, it's then in my and everyone's interest to bring the bug to the attention of the maintainer.
Being able to review the code that actually causes my problem, and having the problem in the first place is a massive leg up.
I've had this problem on projects that I've worked on. We're working in a Delphi project right now. We bought the source to several tools we use, and we've found several bugs, and FIXED them.
Without the source, we would have to interest the vendor in responding to the problem, (really difficult often!) explain the problem, (no easy task) help the vendor replicate it (don't even get me started) and finally get the patch from the vendor.
So OSS and just plain having the source make lots of difference in fixing bugs!
It does look as though unless the Cable Co reprovisioned the cable modems to use another 6Mhz channel, the "usual" limits are ~27Mb/s down and 2-10Mb/s up.
I don't have the brain power right now to do the calcs to see what an acceptable "oversell" rate would be for the number of persons per cable segment. But it does seem that 1000 users for 27Mb/s might be a bit small esp during peak times.
1000 users * 128Kb/s = 128Mb/s? or 500*128Kb/s=64Mb/s
I understand that everyone won't be pulling bandwidth simultaneously, but I can see a fairly consistant pull of 128Kbps for each user avg. Just think of those NNTP/Napster/Redhat ISO hogs pulling their full 1.5+Mb/s - evidently @Home used to cap at 3+Mb/s!
Anyway, it would be great to have some average statistics on cable segment population etc. But you _know_ we'll never see that stuff! [grin] Too much competitive risk to telling us mortals that stuff. Even more interesting would be oversell of IP bandwidth of both cable and DSL.
I suspect oversell in some cases to be more than 100:1 (For each 100 Mb/s DSL/Cable bandwidth to users, the ISP provides 1Mb/s) I think my ISP is better than that, but I woudn't be at all surprised to see some incredible oversell rates. Again, I don't know current Teir 1 T1 Rates (1.544) but I have heard it's around $1000 USD (I just did some research, and that seems about right...) Thus, to break even (not considering overhead and other costs) at $40/month of 256/256 DSL, you would have to sell 40 accounts. That's an oversell of 6.67:1. Since the DL cap for Cable is so much higher, I would expect much higher oversell rates - but they might not effect some users as much (they use during off peak hours etc)
That's a lot of invective, for a guy like you, who must be a perfectly nice guy.:)
I didn't say that DSL service better than Cable service. I took the time to explain the issues, and then concluded that the technical merits of DSL were better than those of cable. (In response to an incorrect statement of DSL vs Cable, when the origional poster was talking about ISP oversell, which isn't exclusive to cable or DSL or dial-up for that matter.
You speak as if changing DSL isps is an easy thing, as if it's likely that you'll be able to find another provider who will service you, who will not have oversold upstream bandwidth
But at least on DSL I get opportunity to choose huh? But I guess that's a bad thing? Again, we talking about technical merits - not what's best for your situation. Think abstract! [Sheesh]
I have switched several DSL lines to new providers, and have virtually no problems. The only change for DSL is a PVC or ATM point. A decent ISP shouldn't have any problem. The telco might screw things up, but that's life I guess.
The basic point is that I like lots of choices, if possible. DSL gives me these choices. ISP/Various data rates/Offer services/Static IP and many others. For some cable will be a better deal.
doesn't clog like you say it should
I didn't say it would clog. We're talking about theoretical use here. If the Cable co puts 10,000 users on your local cable loop, you will have problems. You may be on a loop with only 50 people, and don't have problems. That doesn't mean that it's not possible, or that it will never happen. [By your reasoning, we don't need to worry about some rogue nation H-bombing us, cause it hasn't happened yet]
I guess I'm wasting my breath trying to explain theoretical analysis, but that's the point!
The general point was that DSL is a more stable technical solution. It might not be the most cost effective for you, but it is a better technical solution. What works for you is your choice...
Well, have you ever noticed that No or Cancel isn't always the safe thing to do when asked something by Windows?
For example, when you virus scanner finds a virus, the usual choices are - Delete it, Quarantine it, or Don't do anything. So saying No here is BAD. I can't come up with other examples, but I am sure they are many.
We're not talking about legislating rights! SHEESH We're talking about reasonable OS and application design. A reasonable design will anticipate common user errors, and TRY to make it difficult to make them. Take the chainsaw example. Most chainsaws have a kickback brake. This means when you use the saw improperly, and it kicks back, the saw stops the chain so you don't get turned into dog food. Have we eliminated choice? No, we just did proper engineering to prevent accidents where possible, even when the saw in misused!
As for rights that are gone since 9/11, just thank Seig Heil Ashcroft. The justice department was just waiting for such an event, and then asked for the moon. Unfortunatly, since there we're not any BIG MONEY interests opposing such laws, the individuals of the US got screwed. [Depressing huh?!]
Cool. Now if only the Telco's could be compelled to provide a frame circuit when the location is less than 20,000 feet from the CO _at DSL prices_... (Also force the telco to provide up to at least 1.544 Mb/s)
I guarantee that would speed up the rollout rate for DSL. Fame is lots more costly to roll out, and DSL just siphons off revenue from those frame ciruits. So the telco just doesn't provide DSL where it doesn't want to, or provides a wimpy DSL connection (i.e. 7Mb/s down 2Kb/s up [sheesh])!
But a DSL customer get the opportunity to change this. Cable doesn't. Just switch ISP's. (Not only that, but cable rarely even gives you the choice of ISP's either, so either type of congestion will screw you.)
My ISP charges for bandwidth used. That means that if they keep congestion down, they get the opportunity to bill me more money. That's incentive to keep oversell to a minimum.
Choice is a good thing! DSL isn't perfect, but it does allow for a more choice rich system.
Says who? Cisco's sure isn't, and I don't know anyone else that is.
There are some solutions possible, by have fast changing keys via Radius with Cisco and others, but this only gets you part way there.
I don't trust the current 802.x wep at all. I might consider trusting the "new" standard when it arrives, and gets some serious peer review. Until then, I don't assume that ANY 802.x wep is secure, even trivially so. I assume that I might as well be packet capturing all traffic (in the clear) over said wireless net and handing it off to any anonymous person who requests it.
Cable is shared bandwidth from the cable company CO or central office to the users on that local cable loop segment.
DSL is NOT - it's basically a PPP style connection from the CO to you.
Thus, your won't suffer from performance lag from other users in your segment.
Now, if the DSL is terminated in a ATM or Frame cloud that is saturated, you'll have to compete with those other requests from other customers to the SAME ISP.
If your ISP oversells too much, you can always switch ISP's, and then provided the new ISP doesn't oversell, the problem is fixed.
If you are on the same cable segment (cable modem here) and another group of users saturate the cable segment, the only solution is to blow up those users who are saturating your segment. Since you stand to do a long stint with Bubba in the local pen. for such an act of terrorism (Seig Heil Ashcroft!) it doesn't seem like such a great solution.
For these reasons, DSL has some significant advantages over cable.
So, it would be perfectly OK for GM to put a "change oil" button on the dashboard of all cars. When you push the button, all the oil is dumped out of the engine.
Every year around 250,000 people destroy their cars from this button.
Who's problem is it? The user - probably. But it's GM's problem too. Any problem that occurs in significant numbers is a DESIGN problem. Sure, the user shouldn't, but good design will make it more difficult for a user to screw themselves. It will ALWAYS still be possible - the goal is to make it less likely that the user will do so unwittingly.
Thus, MS's bugs are mostly MS's fault. They don't care about decent design. To blame the user is a cop-out.
Slashdot Mirror
Thanks on the IRQ's thing.
Yes, I just mean that the IRQ interface as it's implimented is way lame.
Put more than a couple RAID/Network etc cards in your server, and poof, all IRQ's are used. Now what?
What I really mean, is to make a "IRQ" system that works right. It's kind of like having 256MB of RAM under windows 3.1, but still running out of GDI/User/System resources - even though lots or RAM is still free...
Cheers!
How about getting rid of IRQ's on the PC platform!
How about getting rid of drive letters in Windows/Dos and having mount points!
How about a better drive interface than the stupid IDE interface. (Macs did it right with SCSI, but now to be "cheap" they do it too [sigh])
And for self healing? If Windows is still around and the predominant OS, I'll pass on the "self healing" - it'll be more like "death-without-dignity." Remember NT 4 SP 6? [Shivver] I don't want MS "self-healing" my machine!
In fact, I don't think I want anyone self healing my machine until software is lots more robust than it is now. At least when I apply patches to my machine and notice that something isn't working right, I know I _just_ patched it, so it might be the patch. With someone else applying patches without my knowing, I would be screwed!
Yeah, all those "wonderful things are just around the corner" articles are neat, but I would truly be happy with some "incremental" changes.
Lets forget "visionary" for a while and just fix the crap that's broken right now! Pleeeeease!
Cheers!
I know it's sorta OT...
Ever notice how the "tort reform" people always point out how some moron who cut his hair with the lawn-mower got 43 gazillion dollars, and thus the system needs reform?
You never hear about the harrassing lawsuits large corporations bring on each other...and in fact, in all the legislation I've heard about, the changes desired were to limit pain & suffering etc. They never limit the ability for big biz to sue and reap huge rewards against each other.
Semms fishy to me. Bah - the rascals!
Cheers!
Thanks for the info. I don't mod & thus no mod points, or I'd give you some...
Thanks again!
Moderators - mod parent up. Excellent info!
Either software is entirely different or it isn't.
Go read some good material on software design. "The Software Conspiracy" by Mark Minasi (I spell horribly, so beware)
Well designed software isn't that much more expensive. Until it gets done regularly, it will be hard to tell how much more expensive it will be, but I can see it potentially costing less.
I don't demand flawless, but then tort law doesn't either. It just demands a reasonable standard. What's reasonable, is what you can convince a jury of...
Software is continually ruling our lives...
Would it be OK, if your car just stopped regularly? It's just software, sure, it controls the fuel/air mixture, but hey it's just electrons running around anyway...
We'll get decent software when we demand that it function "just like everything else other than software." We expect more from the hardware that we run said software on. We wouldn't tolerate cars or microwaves or cell phones or PS2's that run as badly as much software does. That's just plain sad. I guess we get what we deserve.
Until the public starts to see that software design isn't black magic, that no-one can predict, and that it's artistry...blah blah blah, we won't get good software...well that's life I guess.
Cheers!
What is being described is just the very, very, very early, most rudimentary stages of learning about embryos, and the uterus.
_ HELLP.html
One would think from the article, that "in the next few weeks, we'll be growing babies in an acquarium!"
Doctors and Scientists don't even understand some very basic problems in pregnancy in people during normal pregnancy.
Doctors have no real idea what causes HELLP/Pre-Eclampsia. See http://www.dartmouth.edu/~obgyn/mfm/PatientEd/PIH
or
http://www4.ncbi.nlm.nih.gov/PubMed/
search on HELLP
Not only that, they don't know how to stop it, or treat it with any real effectiveness. The only solution, is to hope that the baby is sufficiently developed that it will survive outside the uterus. Then they symptomatically treat the disease. (i.e. Treat blood-pressure, but can't figure out why it goes sky-high, and can't prevent this.) They keep the baby from having problems, and develop the lungs with steriods. The only solution to keep the mother and baby from dying (provided it's mature enough to live outside the uterus) is to deliver the baby.
I know this because I have several OBGYN family members, and my wife had a very serious case of HELLP. It's way serious...and science doesn't have a clue.
Thus, my skeptcism about this breakthrough. Sure, it's incredible science, and shows lots of hard work. But we're so far from actually doing what's proposed, it laughable. It's like giving a press release about your child who just learned to stand-up today, and claiming that they will be challenging Michael Johnson as a sprinter next week. Ha!
I wish them good luck, and kudo's. But just remember they have light-years left to go.
Finally, the whole process brings up a whole host of ethical issues. Lets say, they get it where they can keep embryos alive for many weeks. Do we know they are doing ok, or will they have missed some vital hormone sequence? Getting a baby alive to term is a big enough task, but will we have a normal baby?
What about the babies that reach some form of viability, but have extremely serious health problems. Do we euthanize them?
In short, we only have the faintest inkling of how things work. We haven't solved any of the ethical problems. We've got a lot of work to do before any of this could happen.
I normally end with "Cheers!" but the topic seems way too serious to be flip...
I'm sure you're right, at least in the case you describe.
:) )
The point is, that DSL is a more elegant technical solution. There are fewer problems associated, with large collision domains etc. Just like ethernet was a lousier technical solution than token-ring. Yet for other reasons, ethernet survived. (I think that's a shame...but I digress.
DSL is more easily engineered to not have bandwidth oversell problems, as much as cable is. That doesn't mean that either will succeed. In fact, an on-the-ball cable co might just do it when DSL doesn't. But, just like it's easier to design a good fault tolerant, easy to maintain token-ring network, it's a whole lot easier to design DSL.
Lastly, if I don't have a say in design, which most of us don't, it's a whole lot easier to find a choice that agrees with what we need on DSL. Telco provides wires, nothing else. DSL circuit by someone else, and IP Feed from yet another. As many have pointed out, the IP feed is the most likely culprit, thus, if you can pick ISP's, it's a whole lot easier to solve the IP oversell issue. Thus DSL, is more likely to have an acceptble solution.
The BEST solution is what works...my point is that it seems more likely that DSL will be a preferred solution, especially for exoteric needs.
Cheers!
There was a discussion re: Smoothwall firewall, and the founder/creator, which conincendentially just happens to be Richard (or more appropriately Dick.)
From what I've seen about both guys (it might not be true, but it is from different sources, and widely accepted.) both are abrasive and rather unpleasant. (I'm not saying they're the same, but making some observations, and using both to make a point.)
I've had some friends and acquantences, who thought the whole world was wrong. And in fact, in many ways, they're right! But the way they go about "fixing" the problem - is in itself a problem. They attack everyone. They are abrasive. They are often hypocritical. They burn their bridges, and aren't polite. They use massive shock value to make a point, often offending many, including those who might help them.
Eventually, enough people are really sick of them, and practically want to burn them at the stake (or is that steak?)!
The point is this. I like pleasant people. Pleasant people don't have to compromise their principles, or sell out. But people who are not offensive and have an inflated sence of self importance are just more effective.
RMS probbaly isn't going to change, and that's too bad. From what I can see, his actions are his own worst enemy. Same with Richard from Smoothwall. They could have lots of positive effects, but they make themselves largely ineffective.
That's just too bad.
As to the hypercritical (hyper not hypo) response to them...well what's to surprise? They offend people, and then expect an evenhanded response. Sure, it would be more fair, but I know as well as anyone else, that a fair response is not very likely.
RMS by and large, helps shape the response he gets. I assume that he knows it, and doesn't care. That's the way things go.
Cheers!
Oops...Final sentance.
_It_ would be that (fix the software) or die!
Cheers!
Yeah, I feel sorry for him...
Did you hear that he lost ancestors in the death-camps during the holocaust?
Yup, they fell out of the guard towers...
No, I don't feel sorry for JH. He's just out to grab power for the Justice Dpt. Sure, he probably thinks he's right, but so did Hitler. Thinking you're right doesn't make it so.
I'll be very glad when we're through with him.
Cheers!
Sheesh - do you know what a tort is?
tort Pronunciation Key (tôrt)
n. Law
Damage, injury, or a wrongful act done willfully, negligently, or in circumstances involving strict liability, but not involving breach of contract, for which a civil suit can be brought.
This is what you sue civially for...not loss of life etc. Your definition of liability wouldn't cover my crashing into your car, as long as you didn't die, or have the real possibility of dieing. [Sheesh]
Every, let me say it again! EVERY company should face liability for the products it makes. This is an essential part of a free market system. Everyone is free to make a product, but by participating in a system where everyone is free to create suttf, the maker must ALSO agree to be responsible to a court system.
The court system is where mistakes in the creation of the product are adjudicated. If the maker/creator was negligent in the eyes of a jury, and caused damage to a party, then the creator/maker must "make it as if the damage never occured." Software makers should also suffer this liability.
Now, the hue and cry will be "Oh, software will be so expensive, if we have to eliminate most bugs - software is ALWAYS buggy!" Bah! Software _will_ cost more, if we demand a higher standard. But, that's not such a bad thing. What companies like MS (and others who make shoddy, feature laden, rather than solid software) do is shift the cost burden.
Instead of SuperDuper Wordprocessor costing $800, it only costs (up front) $250. Then the company that buys that software then spends way more fixing problems, and lost documents etc. So, the software really cost $1500 say, but mgmt thinks it only cost $250. Now mgmt should realize this, and probably they eventually will. That will change the system.
But to say that we don't need the legal system is stupid. If I produce a crap lawn-mower that wrecks your lawn, I should be liable. If I produce crap software that tells you to apply 100 Lbs of fertilizer / sqare foot, and that wrecks your lawn, I should be liable. (Free software would probably need some waiver.) What's so different there. There might be some slight modifications needed for software, but if software manufacturers were liable on a "sale of goods" basis for damage from the software, then I expect we would see some serious improvement in the quality of our software. I would be that or die.
Just my thoughts.
Cheers!
Damn straight, and if you don't go out and CONSUME right now, I'll be forced to call Herr Ashcroft on you.
[George says]
Non consumers are EVIL, like that Al Keida guy or guys or something like that...Uhhh, where's Afgainistain again Dick? Nevermind, that stuff makes my head hurt. I'll just take another pretzel!
Yes, I didn't have the "opportunity" to use punch cards... [Groan] You could all see the "old" timer story coming couldn't you...
My prof in university told me of his first programming job. A payroll system. They didn't have a computer system yet, so they diagrammed and setup the program on punch cards. Then they took the completed program (punch-cards) and bought some spare time on another machine. After feeding the punch cards, the program ran correctly tbe first time!
Sheesh, and I ues a compiler as a syntax checker. When was the last time you got anything more complex than "hello world" to compile and run correctly the first time. (Ok, I'm a sucky programmer [grin]) But never the less, program design was a whole lot more rigerous then!
Thems were the days!
Cheers!
Instead of looking at the "needs" of your subordinates, first look at the company you want to work for.
Sure, finding out how to support the people under you is important, but not the most important question.
The most important question, is, "what is the company/mamangement I must work under like?"
If your company is ethical and concerned about it's people (really concerned, not just financially concerned) your job will be much easier. Then the task only becomes finding ways to help your subordinates do their jobs. You'll spend lots less time fighting management above you to actually get this priviledge. That's a huge help.
I know this sounds simplistic, but my exp in this area is that when I am empowered by the employer/upper management, I can really focus on doing what needs to be done. Lots less time is spent on CYA, political fighting, empire building etc. Then you're happy, you can be honest and upfront with your subordinates, and gain their respect and trust. (Trust, i think, is of paramount importance!) Then they'll tell you when you're doing stuff wrong, and help you from looking like a schmuck. Then you can help them get their needs met and be productive.
The end result!? The company runs smoother, more efficiently, and more profitably.
Thus, see what you're empowered to do by your managers, than when it's right, figure out what the specific needs of your subordinates are. They're never the same, but the overall principals are!
Cheers!
But, you fail to see a significant point.
I probably won't go looking through code unless I have a reason. The #1 reason would be a bug/feature that doesn't work right. So, I do a few searches, find no solution. Look at updates etc. still no joy. I crack open the code, review it, run my paticular bug through it, and bingo I find a bug. In OSS, it's then in my and everyone's interest to bring the bug to the attention of the maintainer.
Being able to review the code that actually causes my problem, and having the problem in the first place is a massive leg up.
I've had this problem on projects that I've worked on. We're working in a Delphi project right now. We bought the source to several tools we use, and we've found several bugs, and FIXED them.
Without the source, we would have to interest the vendor in responding to the problem, (really difficult often!) explain the problem, (no easy task) help the vendor replicate it (don't even get me started) and finally get the patch from the vendor.
So OSS and just plain having the source make lots of difference in fixing bugs!
Cheers!
Another cool link - at least to me.
1 02 S0006
Re: Tier 1,2,3 etc ISP costs and considerations...
http://www.networkmagazine.com/article/NMG20011
Cool, thanks!
It does look as though unless the Cable Co reprovisioned the cable modems to use another 6Mhz channel, the "usual" limits are ~27Mb/s down and 2-10Mb/s up.
I don't have the brain power right now to do the calcs to see what an acceptable "oversell" rate would be for the number of persons per cable segment. But it does seem that 1000 users for 27Mb/s might be a bit small esp during peak times.
1000 users * 128Kb/s = 128Mb/s? or 500*128Kb/s=64Mb/s
I understand that everyone won't be pulling bandwidth simultaneously, but I can see a fairly consistant pull of 128Kbps for each user avg. Just think of those NNTP/Napster/Redhat ISO hogs pulling their full 1.5+Mb/s - evidently @Home used to cap at 3+Mb/s!
Anyway, it would be great to have some average statistics on cable segment population etc. But you _know_ we'll never see that stuff! [grin] Too much competitive risk to telling us mortals that stuff. Even more interesting would be oversell of IP bandwidth of both cable and DSL.
I suspect oversell in some cases to be more than 100:1 (For each 100 Mb/s DSL/Cable bandwidth to users, the ISP provides 1Mb/s) I think my ISP is better than that, but I woudn't be at all surprised to see some incredible oversell rates. Again, I don't know current Teir 1 T1 Rates (1.544) but I have heard it's around $1000 USD (I just did some research, and that seems about right...) Thus, to break even (not considering overhead and other costs) at $40/month of 256/256 DSL, you would have to sell 40 accounts. That's an oversell of 6.67:1. Since the DL cap for Cable is so much higher, I would expect much higher oversell rates - but they might not effect some users as much (they use during off peak hours etc)
Anyhow...just interesting stuff..
Thanks again for the Link!
Cheers!
That's a lot of invective, for a guy like you, who must be a perfectly nice guy. :)
I didn't say that DSL service better than Cable service. I took the time to explain the issues, and then concluded that the technical merits of DSL were better than those of cable. (In response to an incorrect statement of DSL vs Cable, when the origional poster was talking about ISP oversell, which isn't exclusive to cable or DSL or dial-up for that matter.
You speak as if changing DSL isps is an easy thing, as if it's likely that you'll be able to find another provider who will service you, who will not have oversold upstream bandwidth
But at least on DSL I get opportunity to choose huh? But I guess that's a bad thing? Again, we talking about technical merits - not what's best for your situation. Think abstract! [Sheesh]
I have switched several DSL lines to new providers, and have virtually no problems. The only change for DSL is a PVC or ATM point. A decent ISP shouldn't have any problem. The telco might screw things up, but that's life I guess.
The basic point is that I like lots of choices, if possible. DSL gives me these choices. ISP/Various data rates/Offer services/Static IP and many others. For some cable will be a better deal.
doesn't clog like you say it should
I didn't say it would clog. We're talking about theoretical use here. If the Cable co puts 10,000 users on your local cable loop, you will have problems. You may be on a loop with only 50 people, and don't have problems. That doesn't mean that it's not possible, or that it will never happen. [By your reasoning, we don't need to worry about some rogue nation H-bombing us, cause it hasn't happened yet]
I guess I'm wasting my breath trying to explain theoretical analysis, but that's the point!
The general point was that DSL is a more stable technical solution. It might not be the most cost effective for you, but it is a better technical solution. What works for you is your choice...
Cheers!
Well, have you ever noticed that No or Cancel isn't always the safe thing to do when asked something by Windows?
For example, when you virus scanner finds a virus, the usual choices are - Delete it, Quarantine it, or Don't do anything. So saying No here is BAD. I can't come up with other examples, but I am sure they are many.
We're not talking about legislating rights! SHEESH We're talking about reasonable OS and application design. A reasonable design will anticipate common user errors, and TRY to make it difficult to make them. Take the chainsaw example. Most chainsaws have a kickback brake. This means when you use the saw improperly, and it kicks back, the saw stops the chain so you don't get turned into dog food. Have we eliminated choice? No, we just did proper engineering to prevent accidents where possible, even when the saw in misused!
As for rights that are gone since 9/11, just thank Seig Heil Ashcroft. The justice department was just waiting for such an event, and then asked for the moon. Unfortunatly, since there we're not any BIG MONEY interests opposing such laws, the individuals of the US got screwed. [Depressing huh?!]
Cheers!
Cool. Now if only the Telco's could be compelled to provide a frame circuit when the location is less than 20,000 feet from the CO _at DSL prices_... (Also force the telco to provide up to at least 1.544 Mb/s)
I guarantee that would speed up the rollout rate for DSL. Fame is lots more costly to roll out, and DSL just siphons off revenue from those frame ciruits. So the telco just doesn't provide DSL where it doesn't want to, or provides a wimpy DSL connection (i.e. 7Mb/s down 2Kb/s up [sheesh])!
Cheers!
Do you have some links to backup this info?
I'm not suggesting you're blowing smoke, but I suspect that the "digital" cable TV signal uses a significant portion of this bandwidth.
Also, how many users are on a local segment.
A few dozen, no problem. 20,000, that's a big problem.
I'd love to see some figures...
Thanks!
Cheers!
But a DSL customer get the opportunity to change this. Cable doesn't. Just switch ISP's. (Not only that, but cable rarely even gives you the choice of ISP's either, so either type of congestion will screw you.)
My ISP charges for bandwidth used. That means that if they keep congestion down, they get the opportunity to bill me more money. That's incentive to keep oversell to a minimum.
Choice is a good thing! DSL isn't perfect, but it does allow for a more choice rich system.
Cheers!
Encryption flash upgradeable?
Says who? Cisco's sure isn't, and I don't know anyone else that is.
There are some solutions possible, by have fast changing keys via Radius with Cisco and others, but this only gets you part way there.
I don't trust the current 802.x wep at all. I might consider trusting the "new" standard when it arrives, and gets some serious peer review. Until then, I don't assume that ANY 802.x wep is secure, even trivially so. I assume that I might as well be packet capturing all traffic (in the clear) over said wireless net and handing it off to any anonymous person who requests it.
Cheers!
[Pinhead alert]
Cable is shared bandwidth from the cable company CO or central office to the users on that local cable loop segment.
DSL is NOT - it's basically a PPP style connection from the CO to you.
Thus, your won't suffer from performance lag from other users in your segment.
Now, if the DSL is terminated in a ATM or Frame cloud that is saturated, you'll have to compete with those other requests from other customers to the SAME ISP.
If your ISP oversells too much, you can always switch ISP's, and then provided the new ISP doesn't oversell, the problem is fixed.
If you are on the same cable segment (cable modem here) and another group of users saturate the cable segment, the only solution is to blow up those users who are saturating your segment. Since you stand to do a long stint with Bubba in the local pen. for such an act of terrorism (Seig Heil Ashcroft!) it doesn't seem like such a great solution.
For these reasons, DSL has some significant advantages over cable.
Cheers!
So, it would be perfectly OK for GM to put a "change oil" button on the dashboard of all cars. When you push the button, all the oil is dumped out of the engine.
Every year around 250,000 people destroy their cars from this button.
Who's problem is it? The user - probably. But it's GM's problem too. Any problem that occurs in significant numbers is a DESIGN problem. Sure, the user shouldn't, but good design will make it more difficult for a user to screw themselves. It will ALWAYS still be possible - the goal is to make it less likely that the user will do so unwittingly.
Thus, MS's bugs are mostly MS's fault. They don't care about decent design. To blame the user is a cop-out.
Cheers!