Or he could go the way of Venezuela and other countries, and declare that instead of saying "no Microsoft" just say "no closed source software". Then Microsoft is free to bring an open source offering to the university, nobody is being 'locked out', and nobody can complain about unfairness.
Congratulations. In your haste to advocate open source, you have just prevented them from using most of the best software in the world. Well done.
God damn, I hate "Use open source, just because" posts.
In that case how about calling the Sales Dept for complaints like these?
That also seems like a good idea, and it's one I've tried in the past. Unfortunately, as soon as they realise you're calling for support, you tend to get transferred to someone in support (or rather, stuffed into the same queue you would have been anyway to wait half an hour for a headset weenie). Sales teams are typically very smooth about this; you'll get "I'll just transfer you to one of my colleagues" and that's it, before you have chance to say a word, you're in the queue. There's not much you can do about it; sales people typically don't have the authority to deal with support issues anyway.
It really depends on their terms of usage agreement [...] If they have a clause in their contract that says that they are not responisble for damage caused to 3rd party modems by their system then they have essentially covered their butts, and he is screwed out of a modem.
<sigh> Would people around here please start to understand that just because a large, profit-making group states something on a piece of paper, it does not put them or their claim above the law. You can say "I'm not responsible for my negligence" as much as you like, but it means jack until a court backs you up on it when asked.
You may be able to accuse them of purposely destroying your modem so that you would have to use the rental, but that would only be valid if they are making you pay extra for the rental.
WTF? If they really did damage his modem, they owe him whatever it takes to restore it to its previous state, or an equivalent replacement. It doesn't matter if they offer him a "free" rental; what if he wants to switch to another service provider in the future, or sell his modem to a third party?
Pick up the phone and dial their support department and calmly ask for a manager.
There are a couple of minor problems with this approach when dealing with most major companies these days.
That support call is probably not free. In fact, for many ISPs, it's premium rate. It'll cost you your day's pay to sit on hold for an hour.
It is standard practice in call centres that everyone is a "supervisor", so when someone asks to be transferred to same, they just get to speak to another headset weenie with no more ability to help them than the first guy. Getting transferred to someone higher up with any real power is difficult.
I wouldn't bother calling phone support for an issue like this. Just send them a registered letter to their customer services manager, stating your problem, what you'd like to have them do about it, and a reasonable time period in which you want them to act. If they don't, maybe send them one reminder letter. Don't get all clever and lawyery if you're not a pro, just make your case in plain English, and ask for a sensible response. Then take them to the small claims court if they fail to respond reasonably.
In open-and-shut cases, which this could well be (though obviously we don't know the full details here), the UK small claims courts are actually pretty good. There's no US-style paying a fortune for a lawyer; in fact, you probably don't need one at all, though you might want to speak to your local Citizen's Advice Bureau. Hearings in the court tend to be brief, and the few results I've seen have been pretty fair to all concerned.
I'm no lawyer, but from past experience, the court would probably award the cost of an equivalent new modem plus court costs, or something similar.
They don't work too well, either. My PC at the office asked me to install two updates earlier this week, for MSXML 2.6 and 3.0. I said OK, and it went away to download and install them.
Next day, I come in, and guess what? New updates are available. Well, "new" as in "the ones you've already got from yesterday", anyway. So I let it do everything again, and it's happy.
You'll never guess what happened this morning...
This is a vanilla Dell PC, nothing silly installed, updated regularly as Windows Update has suggested, and it's confused.
You have control. You have the ability to choose a provider that's capable providing you with the necessary tools to understand your utilization needs and usage. If your provider can't give you access to a bandwidth utilization graph, go find another one.
A bandwidth graph is no good to a small company, with only two full-time IT guys who spend most of their day setting up the PCs that people work on instead of monitoring usage stats in real time. They probably won't even be in the machine room during the time it takes for a serious spike to take effect. And most of your customers are going to be like this if you're a typical large ISP.
I'm not asking my clients for a blank check, I'm asking them to be diligent of their usage.
If you are opening them up to unlimited incoming bandwidth and charging them for it, you are asking them to write you a blank cheque. And I'm afraid "check your data in real time and get back to us immediately if anything is wrong" isn't a viable policy for most users.
If you are not opening them up in that way, then that's great, but then your organisation has little to do with the subject of this thread.
Take an active role in your internet usage and you are largely immune to this sort of billing. [Emphasis added]
Sure. And what about the time someone who doesn't like me sets up a massive attack that spikes at gigabytes of inbound bandwidth within minutes, and over which I have absolutely no control?
You are asking the customer to write you a blank cheque for something about which he can do nothing, no matter how prepared he may be. That is unreasonable, pure and simple.
You, however, can do something about it. As you pointed out yourself:
When one of our circuits gets hit with a Ddos, we call our upstream provider and have them block the attack at their router. We incur no cost for this, it's covered under our contract.
So what's wrong with applying the same principle to your customers? If it's a massive spam attack, you guys have far more resources to detect and deal with it than most of your customers do.
You want the argument to hold from both sides, as long as you're on the winning end of each. That's a great goal from a business standpoint, but it's still not going to win you any awards for logic and finding a solution that is fair to all involved.
We're talking about inbound spikes, over which the customer has no control. They could have a rock solid system with every patch that's ever been invented on it, and it would make no difference.
Whereas with open source software, you have no legal recourse if the latest release of sendmail or bind has an exploit, but rest assured that within 24 hours a fix will be released. Compare that with response times from commercial closed source vendors...
Sure, because it's well known that commercial software vendors never fix serious vulnerabilities as fast as the open source community. Particularly ones like Apple, for example, who have fixed several vulnerabilities in MacOS X way before the equivalent Linux patches were released. Since you like sendmail so much, I suggest you check how fast the major commercial *nix vendors released their patches compared to the open source world, and get back to us.
Now please pick up your ill-informed pro-OS FUD and go away.
Ah, I see. Yes, if you're talking about all graduates, the number quoted is quite credible (though I too suspect rather fewer companies). As anyone who comes into Cambridge on the southbound A14 in the morning can testify, commuting isn't always a good plan, though...:-/
I'd far rather talk to somebody who showed the initiative to send us his resumeé directly than somebody who just sent his CV to a headhunter.
This is a tricky one. I've suffered the same thing: beautifully typeset resume, as provided by me, turned into Word document in Courier with dull and boring written all over it, as provided by agency.
Then again, six months after having my own well-presented CV with covering letter rejected by a potential employer, I got a job at that particular place of employment via a reputable agency. At that point, you have to wonder whether the agencies are actually onto something when they claim to know what their employers want.
Then again, it could just be that the vacancy didn't really exist six months earlier, and did by the time I got serious about looking again and went to the agency. From my now-insider perspective, this seems an equally plausible explanation, and I guess I'll never know...
I read that Cambridge UK employs over 55,000 graduates in 15,000 companies.
Depends on what you mean by "graduates".
It's true that there are lots of small IT companies in and around Cambridge; there are two whole business parks full of them, for a start, one of which is basically a collection of new start-ups just getting off the ground. Quite a few Cambridge University graduates stay on in the area and get jobs with these places; short of getting something in London, the market is probably better in East Anglia than most places in the UK just now.
On the other hand, I find the numbers you gave a bit suspect. Those two business parks I mentioned only house a few hundred companies, and Cambridge isn't a big city, so where are the other 14,000? Also, given that there are two universities in Cambridge, and between them they only have a total annual graduation of a few thousand people, your numbers would suggest that several whole universities' worth of graduates came to Cambridge to work for all those companies that I've never seen. Something doesn't add up somewhere.
Did your figures perhaps refer to Cambridgeshire, the county, rather than just Cambridge, the city? Or maybe "graduates" includes everyone who started working for these companies straight out of uni, even if they've now been there a decade?
Sure, the cylons will learn to shoot... just as soon as a crack commando unit sent to prison for a crime they didn't commit are the lead characters in a series in which no-one dies, in spite of numerous gun-fights in every episode.;-)
I suppose, what the world needs is a law to say that if you send someone a letter threatening legal action if they don't do something - then if they don't do it, you should be REQUIRED to take them to court - and to be liable for their costs, pain & suffering, mental anguish, etc, etc if they turn out to be innocent.
If you think about it, there are already many laws against gaining an advantage via menaces in most places. Things like harrassment, racketeering, blackmail and libel are of dubious legality almost everywhere. What it takes is for them to make one big mistake against someone with the legal might to hit them back with a test case that sets a sufficient precedent, so all the little people can queue right up behind and sue the **** out of them based on that same precedent.
Of course, it helps if, like the UK, the loser of a court case typically winds up paying the costs of both sides. That means if someone really hasn't got a case against you and you know it, you can call their bluff, and it actually costs them money to look stupid, while costing you nothing but a little time. Alas, I gather the US legal system does not work this way; perhaps it should...
I was wondering if anyone was going to pick up on the Angel reference. I've never really seen it much myself -- it was on satellite TV way before terrestrial, and I only have the latter -- but here in the UK, several of my friends (and we're a load of Buffy fans) rate the first few series of Angel higher than the last few of BTVS.
"Most scripting languages are designed around
letting small problems be implemented quickly."
Isn't that the core philosophy of Microsoft's Windows Update service?
No, I think you'll find it's the core philosophy of their development groups, which are capable of generating small (and even large) problems at a faster rate than almost anyone else on the planet.
The Windows Update Service is to fix those small problems slowly later... particularly if you're a 56K modem user.
Yep, serious programs need to be storing all strings in some sort of resource of the form "blah blah {1} blah {2} blah blah blah {3} blah", where the {n} are ordered replaceable parameters such as numbers or dates. A translation agency can deal with this, adjusting the word order and idioms as appropriate and giving you back a string in the same format, with the same markers for replaceables (though not necessarily in the same order).
This is kinda what printf should have been, but you can't change that order straightforwardly in a printf format string. This is also the single biggest reason that C++ IOStreams suck for real world applications aimed at multiple markets. The idea of extensibility via chaining things together with << was nice, but they forgot that the chaining has to be soft-coded so you can translate without changing the code, not hard-coded in the app.
A verifiable scientific study is =impossible= because closed source is, by definition, closed. This is the best we're going to get.
Not to everyone. If they at least published details of the alternatives in question, those working on them could dispute the claims if they wished to do so, or others could attempt to reproduce the claimed problems (which doesn't require the source, of course) to validate the findings.
No tool picks up all errors and all kinds of errors.
Of course not. But pretty much any good automated tool of the kind in question picks up all of the types mentioned. The fact that major OS vendors and those responsible for the Linux distributions don't run such things as a matter of course is pretty sad.
Small? You've never worked on a stack, have you?
Actually, yes, I've spent months working on code with a similar purpose and complexity, though admittedly never specifically a TCP/IP stack. You don't know how big a large software project is, though, do you?:-)
I found this whole thing a bit suspicious from the start, and my initial reaction was much th same as KefkaFloyd's.
For the RTFA-impaired, here are some choice quotes from the article, and my take on them.
"Reasoning declined to disclose which operating systems it compared with Linux, but said two of the three general-purpose operating systems were versions of Unix."
IE: This is not in any way a verifiable scientific study, it's a means to attract people to our web site.
"For the comparison products, the company had access to the source code that for proprietary software is usually a closely guarded secret."
IE: More of the same.
"Reasoning looked for programming problems such as memory that was marked as free when it was in fact still in use, memory that was being used without being properly initialized and attempts to store data that exceeded the space reserved for it."
IE: They tested for basic mechanical errors. It's sad that production code should have these, but anyone could run tools (probably just like the one these guys sell) to fix them. If your process includes automated runs with Lint, Purify, etc. then you should never suffer from these at all. It's actually a pretty damning indictment of all concerned, including Linux, that any such bugs featured in a piece of code as small as the TCP/IP stuff for an OS.
And best of all...
"Trappe said his company didn't measure the comparative performance of the different versions TCP/IP, something that would have been difficult because of hardware differences such as network acceleration hardware on the network-specific products."
IE: The playing field was not level. For all we know, they compared state-of-the-art yet-to-be-widely-seen algorithms from next generation OSes to tried and tested from Linux.
Hell, for all we know, one of the big Linux distributors commissioned the survey in the first place.
And yeah, I gave up at the point when they wanted all my contact details with no statement about how they'd be used, too.
Slashdot Mirror
Congratulations. In your haste to advocate open source, you have just prevented them from using most of the best software in the world. Well done.
God damn, I hate "Use open source, just because" posts.
That also seems like a good idea, and it's one I've tried in the past. Unfortunately, as soon as they realise you're calling for support, you tend to get transferred to someone in support (or rather, stuffed into the same queue you would have been anyway to wait half an hour for a headset weenie). Sales teams are typically very smooth about this; you'll get "I'll just transfer you to one of my colleagues" and that's it, before you have chance to say a word, you're in the queue. There's not much you can do about it; sales people typically don't have the authority to deal with support issues anyway.
<sigh> Would people around here please start to understand that just because a large, profit-making group states something on a piece of paper, it does not put them or their claim above the law. You can say "I'm not responsible for my negligence" as much as you like, but it means jack until a court backs you up on it when asked.
WTF? If they really did damage his modem, they owe him whatever it takes to restore it to its previous state, or an equivalent replacement. It doesn't matter if they offer him a "free" rental; what if he wants to switch to another service provider in the future, or sell his modem to a third party?
There are a couple of minor problems with this approach when dealing with most major companies these days.
I wouldn't bother calling phone support for an issue like this. Just send them a registered letter to their customer services manager, stating your problem, what you'd like to have them do about it, and a reasonable time period in which you want them to act. If they don't, maybe send them one reminder letter. Don't get all clever and lawyery if you're not a pro, just make your case in plain English, and ask for a sensible response. Then take them to the small claims court if they fail to respond reasonably.
In open-and-shut cases, which this could well be (though obviously we don't know the full details here), the UK small claims courts are actually pretty good. There's no US-style paying a fortune for a lawyer; in fact, you probably don't need one at all, though you might want to speak to your local Citizen's Advice Bureau. Hearings in the court tend to be brief, and the few results I've seen have been pretty fair to all concerned.
I'm no lawyer, but from past experience, the court would probably award the cost of an equivalent new modem plus court costs, or something similar.
They don't work too well, either. My PC at the office asked me to install two updates earlier this week, for MSXML 2.6 and 3.0. I said OK, and it went away to download and install them.
Next day, I come in, and guess what? New updates are available. Well, "new" as in "the ones you've already got from yesterday", anyway. So I let it do everything again, and it's happy.
You'll never guess what happened this morning...
This is a vanilla Dell PC, nothing silly installed, updated regularly as Windows Update has suggested, and it's confused.
The prosecution rests, your honour.
A bandwidth graph is no good to a small company, with only two full-time IT guys who spend most of their day setting up the PCs that people work on instead of monitoring usage stats in real time. They probably won't even be in the machine room during the time it takes for a serious spike to take effect. And most of your customers are going to be like this if you're a typical large ISP.
If you are opening them up to unlimited incoming bandwidth and charging them for it, you are asking them to write you a blank cheque. And I'm afraid "check your data in real time and get back to us immediately if anything is wrong" isn't a viable policy for most users.
If you are not opening them up in that way, then that's great, but then your organisation has little to do with the subject of this thread.
Sure. And what about the time someone who doesn't like me sets up a massive attack that spikes at gigabytes of inbound bandwidth within minutes, and over which I have absolutely no control?
You are asking the customer to write you a blank cheque for something about which he can do nothing, no matter how prepared he may be. That is unreasonable, pure and simple.
You, however, can do something about it. As you pointed out yourself:
So what's wrong with applying the same principle to your customers? If it's a massive spam attack, you guys have far more resources to detect and deal with it than most of your customers do.
You want the argument to hold from both sides, as long as you're on the winning end of each. That's a great goal from a business standpoint, but it's still not going to win you any awards for logic and finding a solution that is fair to all involved.
We're talking about inbound spikes, over which the customer has no control. They could have a rock solid system with every patch that's ever been invented on it, and it would make no difference.
Someone here has misunderstood open source.
If you think it means that any bug anyone ever sees will get fixed, it's you, though.
In the same vein as the parent, but much more black and white: who paid for the study in question?
Sure, because it's well known that commercial software vendors never fix serious vulnerabilities as fast as the open source community. Particularly ones like Apple, for example, who have fixed several vulnerabilities in MacOS X way before the equivalent Linux patches were released. Since you like sendmail so much, I suggest you check how fast the major commercial *nix vendors released their patches compared to the open source world, and get back to us.
Now please pick up your ill-informed pro-OS FUD and go away.
S'OK, there won't be any games that take advantage of that for months, either, and yours was probably waaaaay cheaper. :-)
Just be glad you didn't spend half as much again on a Geforce FX...
Ah, I see. Yes, if you're talking about all graduates, the number quoted is quite credible (though I too suspect rather fewer companies). As anyone who comes into Cambridge on the southbound A14 in the morning can testify, commuting isn't always a good plan, though... :-/
This is a tricky one. I've suffered the same thing: beautifully typeset resume, as provided by me, turned into Word document in Courier with dull and boring written all over it, as provided by agency.
Then again, six months after having my own well-presented CV with covering letter rejected by a potential employer, I got a job at that particular place of employment via a reputable agency. At that point, you have to wonder whether the agencies are actually onto something when they claim to know what their employers want.
Then again, it could just be that the vacancy didn't really exist six months earlier, and did by the time I got serious about looking again and went to the agency. From my now-insider perspective, this seems an equally plausible explanation, and I guess I'll never know...
Depends on what you mean by "graduates".
It's true that there are lots of small IT companies in and around Cambridge; there are two whole business parks full of them, for a start, one of which is basically a collection of new start-ups just getting off the ground. Quite a few Cambridge University graduates stay on in the area and get jobs with these places; short of getting something in London, the market is probably better in East Anglia than most places in the UK just now.
On the other hand, I find the numbers you gave a bit suspect. Those two business parks I mentioned only house a few hundred companies, and Cambridge isn't a big city, so where are the other 14,000? Also, given that there are two universities in Cambridge, and between them they only have a total annual graduation of a few thousand people, your numbers would suggest that several whole universities' worth of graduates came to Cambridge to work for all those companies that I've never seen. Something doesn't add up somewhere.
Did your figures perhaps refer to Cambridgeshire, the county, rather than just Cambridge, the city? Or maybe "graduates" includes everyone who started working for these companies straight out of uni, even if they've now been there a decade?
It's true, there were exceptions (plural, IIRC) to the rule, particularly in the later series. But usually, they couldn't shoot for toffee...
In fact, if you talked to the production guys, it really was a Cylon car; they were the inspiration for the sliding red light.
Sure, the cylons will learn to shoot... just as soon as a crack commando unit sent to prison for a crime they didn't commit are the lead characters in a series in which no-one dies, in spite of numerous gun-fights in every episode. ;-)
If you think about it, there are already many laws against gaining an advantage via menaces in most places. Things like harrassment, racketeering, blackmail and libel are of dubious legality almost everywhere. What it takes is for them to make one big mistake against someone with the legal might to hit them back with a test case that sets a sufficient precedent, so all the little people can queue right up behind and sue the **** out of them based on that same precedent.
Of course, it helps if, like the UK, the loser of a court case typically winds up paying the costs of both sides. That means if someone really hasn't got a case against you and you know it, you can call their bluff, and it actually costs them money to look stupid, while costing you nothing but a little time. Alas, I gather the US legal system does not work this way; perhaps it should...
I was wondering if anyone was going to pick up on the Angel reference. I've never really seen it much myself -- it was on satellite TV way before terrestrial, and I only have the latter -- but here in the UK, several of my friends (and we're a load of Buffy fans) rate the first few series of Angel higher than the last few of BTVS.
No, I think you'll find it's the core philosophy of their development groups, which are capable of generating small (and even large) problems at a faster rate than almost anyone else on the planet.
The Windows Update Service is to fix those small problems slowly later... particularly if you're a 56K modem user.
Yep, serious programs need to be storing all strings in some sort of resource of the form "blah blah {1} blah {2} blah blah blah {3} blah", where the {n} are ordered replaceable parameters such as numbers or dates. A translation agency can deal with this, adjusting the word order and idioms as appropriate and giving you back a string in the same format, with the same markers for replaceables (though not necessarily in the same order).
This is kinda what printf should have been, but you can't change that order straightforwardly in a printf format string. This is also the single biggest reason that C++ IOStreams suck for real world applications aimed at multiple markets. The idea of extensibility via chaining things together with << was nice, but they forgot that the chaining has to be soft-coded so you can translate without changing the code, not hard-coded in the app.
Not to everyone. If they at least published details of the alternatives in question, those working on them could dispute the claims if they wished to do so, or others could attempt to reproduce the claimed problems (which doesn't require the source, of course) to validate the findings.
Of course not. But pretty much any good automated tool of the kind in question picks up all of the types mentioned. The fact that major OS vendors and those responsible for the Linux distributions don't run such things as a matter of course is pretty sad.
Actually, yes, I've spent months working on code with a similar purpose and complexity, though admittedly never specifically a TCP/IP stack. You don't know how big a large software project is, though, do you? :-)
I found this whole thing a bit suspicious from the start, and my initial reaction was much th same as KefkaFloyd's.
For the RTFA-impaired, here are some choice quotes from the article, and my take on them.
IE: This is not in any way a verifiable scientific study, it's a means to attract people to our web site.
IE: More of the same.
IE: They tested for basic mechanical errors. It's sad that production code should have these, but anyone could run tools (probably just like the one these guys sell) to fix them. If your process includes automated runs with Lint, Purify, etc. then you should never suffer from these at all. It's actually a pretty damning indictment of all concerned, including Linux, that any such bugs featured in a piece of code as small as the TCP/IP stuff for an OS.
And best of all...
IE: The playing field was not level. For all we know, they compared state-of-the-art yet-to-be-widely-seen algorithms from next generation OSes to tried and tested from Linux.
Hell, for all we know, one of the big Linux distributors commissioned the survey in the first place.
And yeah, I gave up at the point when they wanted all my contact details with no statement about how they'd be used, too.