The class of WAN accelerator I was referring to stores massive MD5-indexed MRU caches of frequently transferred datablocks. This definitionally requires an aggregator... a device of some sort. These devices help little if there is not lots of duplicate traffic, but most organizations have very large volumes of duplicate traffic.
I'll have to disagree; as an interviewer I'm interested in hiring a man, not a toolset.
That might be illegal. *cha cha cha*
Anyway, I understand your point. I would hone in on the "good programming instincts" line of yours as worthy of merit. My point with the google/code test is that one could then look at code produced and have a discussion about it. I.e., if they ripped some shitty code and say during your interview, "oh look at this shitty code here, if fails to check errno for E_AGAIN, there's about a one in one billion chance this program will malfunction under load," you will have learned something extraordinary about your candidate. You can do this a variety of ways.
But jump one level up to the OP. This person is conducting interviews that are resulting in lousy job performers, and doesn't quite understand why. I would say that it's because he's either scaring away the good candidates or just failing to recognize them. The job of a good interviewer is to elicit the candidates strengths. OP's position sounds to me decidedly like he believes its his job to demonstrate the candidates weaknesses. It's no wonder that he's having candidates walk out on his company interviews.
I admit, I feel that I've never mastered the magic of doing full due dilligence with a candidate. The only thing that I've had incredible success with is this: if they have open source or other at-home coding and technical hobbies, they are always winners. What's that, though? A 1 in 20 event?
"More than half (55%) of engineering doctorates were awarded to students on temporary visas".
This does not establish a shortage. It establishes that universities accept lots of foreign candidates into their PhD programs, and that lots of foreigners apply. This should be no surprise, these students have to pay full price, and the universities are happy to put forward sufficient educational staff in order to instruct them. I am sorry: but this is not a good argument in favor of a shortage.
There is a strong argument against the argument that there is a shortage. In markets where there is a lack of a supply in the face of consistent demand, the cost of the supplied item tends to go up. If there is a very strong shortage in a specific specialty, one will therefore note significant wage inflation for the speciality. I don't know what PhD specialty you are referring to, but I believe that for CS there is no such inflation.
I used to interview like that, too. I don't think it served me well. Better for you to give a one hour assignment, hooked up to the internet, and walk away for an hour. Don't hover. Give them a problem where Google will help, but not have an exact answer. Let them Google. Googling something is part of the toolset. You're attempting to constrain their toolset. Don't do that.
Part of the problem with this sort of interview is that it expects/declarative/ knowledge. You will find, however, that there are many good software engineers who do quite well with/procedural/ knowledge, and lack finesse in terminology. Some people simply won't commit things you expect to memory; they'll know that it's so easy to remind themselves with their own examples and what not, they be more involved in things other than what you ask.
And as you mentioned, people tend to crumble incredibly under this sort of on the spot test.
American workers aren't investing in R&D education at least in part because of the low wages to be expected from it. Importing foreigners or outsourcing the work, on the grounds that there aren't enough PhD's is a bit of a lie. Rather, there aren't enough CHEAP PhD's. That's what this has always been about.
So it would appear that I was inadvertently exaggerating.:) Be that as it may, 3 years of not earning pay during the early part of your life is a sacrifice. Supply and demand and what not. I'd be curious how socialized medicine handles doctor pay for specialties. I would suppose some forms of specialties are extrinsic to the system (e.g., elective plastic surgery).
One thing that many people miss is that we already have a socialized medical system in the US (of a sort), it's just royally fucked up. Show up at any ER without health insurance: they'll treat you (for some things). The consequences of this forced donation of the hospital are inconsistent and penalize hospitals in poor communities particularly severely. It's capricious, unjust, confiscatory, and inequal. The state should pay, not force hospitals to pay. That's honest, and frankly more transparent. As is, there is a positive cost to "hidden" socialized medicine in the US that is quite difficult to quantify... hell, it's almost socially invisible. Not good.
Guess I would have remembered that, if I'd thought about it (GP itself a residency-specialty). Still GP is a lightweight residency. Any significant residency, and this would be a real negative, producing a truly significant lifetime decrease in income. That being the case, it will result in a significant decrease in folks finishing the residency. The residency will have to reduce its length and special training to compensate, I would think. Image neurosurgeon, where residency can be half of a working career (10-12 years not uncommon). I suppose that one could compensate for the financial loss by paying the neurosurgery RESIDENTS what a working GP would make. Is that what they do?
Another thing social healthcare dose is pay 1st rate GPs the same as 1st rate specialists.
Interesting stuff. See, it takes a couple extra years personal investment in the medical education system to become the least of specialists, and for some types of specialists takes an additional 10 years. So, what you are proposing is a system where the more difficult the speciality, the less the lifetime pay. Strange.
I don't think you're dreaming. I think that discovery may soon look to be a frightening process for the RIAA. We know that they have been using their cases and settlement requests as a glorified blackmail system. The law frowns on that. Once the threat of putting workers at the settlement center under oath becomes a pending reality, we'll see what happens here. Alas, I bet it won't happen. Ironically, this case has "settlement" written all over it.
I'm happy for the RIAA to get anything and everything that's coming to them, but I don't think it will change their litigation-happy behaviour at all.
Depends on the findings. There was a case that came up just recently where the RIAA is being accused of both federal and state RICO charges as well as quite a number of other things. While this is only a civil case, these are criminal charges. If the findings go to the plaintiff on some of the more serious charges, and were the RIAA not to reform itself, the remedies in a follow-on case from any other party could be quite severe indeed.
It's not that offshore people are inherently inferior. It's that most offshore technical resources have little or no interest in technology. They simply want to make money.
You've just described the lionshare of my peers. I'd say that describes 85% of American tech workers.
The rare bird who is passionately in love with his field is extraordinary in any country.
Have you had trouble with the Juniper devices? I'd be interested in some kind of reference to a report. A group of ours has charter to investigate them, have gotten a pair. I doubt their testing will be particularly thorough on that level, so I'd love to hear more. BTW, it was my understanding that the Juniper device was supposed to be indeed "completely transparent." I.e., only mutations that can possibly be detected are timing. There's no theoretical reason why this oughtn't be.
Honestly, I don't know what all the hooplah is about. We've known for ages about the impact of latency and its interplay with windowing on thruputs.
I'm more excited by the new class of devices that are acting as block-caches for generalized TCP/IP traffic. These things are really neat-o for large distributed organziations, because they really help with duplicate traffic of which there is typically LOTS in every large organization. Idea is very simple:
Devices sit at each WAN end point, transparently inserted between ordinary TCP/IP end points. Devices layer the protocols, with traffic sort of like this: "hey, I'm about to send you MD5-such-n-such. Need it. No? Okay, consider it sent." Wash, rinse, repeat. I call this "content-aware compression," (which I admit is a misnomer of sorts), but it can really, reallllly compress some organizations WAN traffic pretty well.
Well. Unless they somehow evolved under very low gravity, although I can see how such an environment might very well not be conducive to "evolving" at all. But supposing that it were, then perhaps they'd look like Octopuses... or, thinking on it, just plain anything. The Ocean has a great deal more variation in its life forms than do our above water areas....
A lesson everyone should learn before attempting to engage coercive forces against a criminal is that the criminal himself is criminally-minded and quite possibly might turn his criminal mind towards you. One should plan for this in advance... with malice aforethought if you get my drift.
Do you imagine that such a company, or the managers involved, will not protect their interests by silencing and punishing someone who speaks out against them?
As you can see comparing and contrasting the two sentences, it should be apparent that of course I would not imagine that.
This discussion is becoming far to much about one person's personal circumstance in life.
...it simply takes your supervisor, or the paperwork pushers, to "express concern" and start skewing employee evaluations...
Be that as it may, this has nothing to do with your written NDA and everything to do with unethical criminal conduct. For this to be an argument about what an NDA can or cannot legally do, it has to be a discussion about the overt behavior of the company. I.E., can they say to a judge, 'look here, we have this NDA that says he cannot report crimes; he did, we fired him for cause, and report it to all of his reference calls'. No. No, they can't.
What those acting outside of the law elect to do is apparently a painful subject to you personally, and I am sorry to hear it.
If this were before you pulled whatever trigger you pulled, I would have advised you to seriously CYA. A lesson everyone should learn before attempting to engage coercive forces against a criminal is that the criminal himself is criminally-minded and quite possibly might turn his criminal mind towards you. One should plan for this in advance... with malice aforethought if you get my drift.
BTW, you intimated previously that you have reference pollution. There are sharks who specialize in that. It's quite a bit easier to develop a case than you might think.
Most don't NDA's don't outright *forbid* such behavior, but they most certainly prevent it. Look again, carefully, at the details of a typical NDA. Reporting the criminal activity will get you fired under the NDA, and leave you in the position to establish the criminal behavior as a defense against an unlawful firing. Unfortunately, you no longer have a job. Nor are you likely to get one when this shows up on your resume.
In many states, it can itself be a crime to not report a felony that you witness. To fire someone for cause for a good faith report on a criminal activity would be highly actionable in every state I'm quite sure,... not to mention act as probable cause for an investigation into the company itself.
As for the presence or absence of a network of undermining HR demons in some companies, I won't comment, except to suggest that sharks can swim on both side of that fence. Plenty of blood to be spilled here; no big company can count on mere HR mortals to perjure themselves on the stand, big risk even taking that one to court. If that's indeed what's happening, and they know it, they will settle. Getting caught in perjury and criminal misbehavior in an employment setting will cause a large corporation enormous damages and is total suicide for smaller companies.
haha, what the fuck are they waiting for legal approval for?
Only people in the company who can release the company's intellectual property are its authorized agents. These agents are typically a company executive, who will have fiduciary for the company, and will need to move as carefully as possible. At a software company, one ought to be particularly careful of the company's intellectual property.
The GPL doesn't require that you release your source code; rather it requires that either you do so or face the possible consequences that copyright law implies. There's the larger (I say, larger) issue of community consequences... that's what SWSoft is facing in the immediate. They should at least make a statement of some kind.
Modsim is a unusual use case. That use case has its own concerns.
20% is really on the high end for utilization in virtually every data center.
The syndrome that is most alive today is the "one service, one box" issue. That's why all the drive for consolidation, coming from the virtualization vendors.
One of my personal points about the 2nd is that it clearly meant, then, what liberals don't want it to mean today. The libertarian interpretation is consistent with the first 150 years of its interpretation. Now, my point: it's fine if the people feel strongly enough to Amend it to mean something less all encompassing today. But the people HAVEN'T DONE THAT. And to have a system of government whereby things you don't like can be stricken due to politics and not procedure means that you really don't have the Constitutional system we were given at all.
Not only was it only enumerated powers, they felt so strongly about it they passed an Amendment just to make it clear they meant only enumerated powers!
20 Dell 1955 Blades; 16G ram; 70GB SAS 10K drive (one) on which ESX 3.0 is hosted (or a variety of Xen flavors); four gigabit ethernet controllers per blade; CISCO 4948 48 port switch, with 4 ethernets per blade bonded; CISCO 6504e core with Sup-32; Net App 3020 and 3050 for NFS and iSCSI; some EMC Clarion units, likewise.
For CPU we used SPEC CPU 2006 and score about 5-6ish % on VMWare as the same test done on those blades in hard metal. Xen is undiscernably different to the subjective eye than hard metal. I would have to break out large batch testing methodology and run the results through inferential statistics to conclude that there was a difference at all.
I/O is a different story.
The Xen performance claims and the VZ performance claims aren't really useful. They're theoretical. As in, "theoretically, we can stack 100 operating systems on this blade efficiently." Think about that. That's just plain nuts. I can't think of a real use case for that.
BTW, if you like OpenVZ, and have the right use case, the commercial Virtuozzo product ranks as the "best virtualization technology that no one has ever heard of" in my book. They really have their IT management story down pat.
Slashdot Mirror
The class of WAN accelerator I was referring to stores massive MD5-indexed MRU caches of frequently transferred datablocks. This definitionally requires an aggregator... a device of some sort. These devices help little if there is not lots of duplicate traffic, but most organizations have very large volumes of duplicate traffic.
C//
I'll have to disagree; as an interviewer I'm interested in hiring a man, not a toolset.
That might be illegal. *cha cha cha*
Anyway, I understand your point. I would hone in on the "good programming instincts" line of yours as worthy of merit. My point with the google/code test is that one could then look at code produced and have a discussion about it. I.e., if they ripped some shitty code and say during your interview, "oh look at this shitty code here, if fails to check errno for E_AGAIN, there's about a one in one billion chance this program will malfunction under load," you will have learned something extraordinary about your candidate. You can do this a variety of ways.
But jump one level up to the OP. This person is conducting interviews that are resulting in lousy job performers, and doesn't quite understand why. I would say that it's because he's either scaring away the good candidates or just failing to recognize them. The job of a good interviewer is to elicit the candidates strengths. OP's position sounds to me decidedly like he believes its his job to demonstrate the candidates weaknesses. It's no wonder that he's having candidates walk out on his company interviews.
I admit, I feel that I've never mastered the magic of doing full due dilligence with a candidate. The only thing that I've had incredible success with is this: if they have open source or other at-home coding and technical hobbies, they are always winners. What's that, though? A 1 in 20 event?
C//
"More than half (55%) of engineering doctorates were awarded to students on temporary visas".
This does not establish a shortage. It establishes that universities accept lots of foreign candidates into their PhD programs, and that lots of foreigners apply. This should be no surprise, these students have to pay full price, and the universities are happy to put forward sufficient educational staff in order to instruct them. I am sorry: but this is not a good argument in favor of a shortage.
There is a strong argument against the argument that there is a shortage. In markets where there is a lack of a supply in the face of consistent demand, the cost of the supplied item tends to go up. If there is a very strong shortage in a specific specialty, one will therefore note significant wage inflation for the speciality. I don't know what PhD specialty you are referring to, but I believe that for CS there is no such inflation.
Therefore there is no shortage, QED.
C//
I used to interview like that, too. I don't think it served me well. Better for you to give a one hour assignment, hooked up to the internet, and walk away for an hour. Don't hover. Give them a problem where Google will help, but not have an exact answer. Let them Google. Googling something is part of the toolset. You're attempting to constrain their toolset. Don't do that.
/declarative/ knowledge. You will find, however, that there are many good software engineers who do quite well with /procedural/ knowledge, and lack finesse in terminology. Some people simply won't commit things you expect to memory; they'll know that it's so easy to remind themselves with their own examples and what not, they be more involved in things other than what you ask.
Part of the problem with this sort of interview is that it expects
And as you mentioned, people tend to crumble incredibly under this sort of on the spot test.
C//
American workers aren't investing in R&D education at least in part because of the low wages to be expected from it. Importing foreigners or outsourcing the work, on the grounds that there aren't enough PhD's is a bit of a lie. Rather, there aren't enough CHEAP PhD's. That's what this has always been about.
C//
I remember GP being 2 years, but apparently not so:
/ 3EDD4E91945F8A2B86256F850071AE49?OpenDocument
:) Be that as it may, 3 years of not earning pay during the early part of your life is a sacrifice. Supply and demand and what not. I'd be curious how socialized medicine handles doctor pay for specialties. I would suppose some forms of specialties are extrinsic to the system (e.g., elective plastic surgery).
Here ya go.
http://residency.wustl.edu/medadmin/resweb.nsf/WV
So the spectrum seems to be 3-6.
So it would appear that I was inadvertently exaggerating.
One thing that many people miss is that we already have a socialized medical system in the US (of a sort), it's just royally fucked up. Show up at any ER without health insurance: they'll treat you (for some things). The consequences of this forced donation of the hospital are inconsistent and penalize hospitals in poor communities particularly severely. It's capricious, unjust, confiscatory, and inequal. The state should pay, not force hospitals to pay. That's honest, and frankly more transparent. As is, there is a positive cost to "hidden" socialized medicine in the US that is quite difficult to quantify... hell, it's almost socially invisible. Not good.
C//
Guess I would have remembered that, if I'd thought about it (GP itself a residency-specialty). Still GP is a lightweight residency. Any significant residency, and this would be a real negative, producing a truly significant lifetime decrease in income. That being the case, it will result in a significant decrease in folks finishing the residency. The residency will have to reduce its length and special training to compensate, I would think. Image neurosurgeon, where residency can be half of a working career (10-12 years not uncommon). I suppose that one could compensate for the financial loss by paying the neurosurgery RESIDENTS what a working GP would make. Is that what they do?
C//
Another thing social healthcare dose is pay 1st rate GPs the same as 1st rate specialists.
Interesting stuff. See, it takes a couple extra years personal investment in the medical education system to become the least of specialists, and for some types of specialists takes an additional 10 years. So, what you are proposing is a system where the more difficult the speciality, the less the lifetime pay. Strange.
C//
I don't think you're dreaming. I think that discovery may soon look to be a frightening process for the RIAA. We know that they have been using their cases and settlement requests as a glorified blackmail system. The law frowns on that. Once the threat of putting workers at the settlement center under oath becomes a pending reality, we'll see what happens here. Alas, I bet it won't happen. Ironically, this case has "settlement" written all over it.
C//
I'm happy for the RIAA to get anything and everything that's coming to them, but I don't think it will change their litigation-happy behaviour at all.
Depends on the findings. There was a case that came up just recently where the RIAA is being accused of both federal and state RICO charges as well as quite a number of other things. While this is only a civil case, these are criminal charges. If the findings go to the plaintiff on some of the more serious charges, and were the RIAA not to reform itself, the remedies in a follow-on case from any other party could be quite severe indeed.
C//
It's not that offshore people are inherently inferior. It's that most offshore technical resources have little or no interest in technology. They simply want to make money.
You've just described the lionshare of my peers. I'd say that describes 85% of American tech workers.
The rare bird who is passionately in love with his field is extraordinary in any country.
C//
Have you had trouble with the Juniper devices? I'd be interested in some kind of reference to a report. A group of ours has charter to investigate them, have gotten a pair. I doubt their testing will be particularly thorough on that level, so I'd love to hear more. BTW, it was my understanding that the Juniper device was supposed to be indeed "completely transparent." I.e., only mutations that can possibly be detected are timing. There's no theoretical reason why this oughtn't be.
C//
Honestly, I don't know what all the hooplah is about. We've known for ages about the impact of latency and its interplay with windowing on thruputs.
I'm more excited by the new class of devices that are acting as block-caches for generalized TCP/IP traffic. These things are really neat-o for large distributed organziations, because they really help with duplicate traffic of which there is typically LOTS in every large organization. Idea is very simple:
Devices sit at each WAN end point, transparently inserted between ordinary TCP/IP end points. Devices layer the protocols, with traffic sort of like this: "hey, I'm about to send you MD5-such-n-such. Need it. No? Okay, consider it sent." Wash, rinse, repeat. I call this "content-aware compression," (which I admit is a misnomer of sorts), but it can really, reallllly compress some organizations WAN traffic pretty well.
C//
Well. Unless they somehow evolved under very low gravity, although I can see how such an environment might very well not be conducive to "evolving" at all. But supposing that it were, then perhaps they'd look like Octopuses... or, thinking on it, just plain anything. The Ocean has a great deal more variation in its life forms than do our above water areas....
C//
A lesson everyone should learn before attempting to engage coercive forces against a criminal is that the criminal himself is criminally-minded and quite possibly might turn his criminal mind towards you. One should plan for this in advance... with malice aforethought if you get my drift.
Do you imagine that such a company, or the managers involved, will not protect their interests by silencing and punishing someone who speaks out against them?
As you can see comparing and contrasting the two sentences, it should be apparent that of course I would not imagine that.
This discussion is becoming far to much about one person's personal circumstance in life.
Later.
C//
...it simply takes your supervisor, or the paperwork pushers, to "express concern" and start skewing employee evaluations...
Be that as it may, this has nothing to do with your written NDA and everything to do with unethical criminal conduct. For this to be an argument about what an NDA can or cannot legally do, it has to be a discussion about the overt behavior of the company. I.E., can they say to a judge, 'look here, we have this NDA that says he cannot report crimes; he did, we fired him for cause, and report it to all of his reference calls'. No. No, they can't.
What those acting outside of the law elect to do is apparently a painful subject to you personally, and I am sorry to hear it.
If this were before you pulled whatever trigger you pulled, I would have advised you to seriously CYA. A lesson everyone should learn before attempting to engage coercive forces against a criminal is that the criminal himself is criminally-minded and quite possibly might turn his criminal mind towards you. One should plan for this in advance... with malice aforethought if you get my drift.
BTW, you intimated previously that you have reference pollution. There are sharks who specialize in that. It's quite a bit easier to develop a case than you might think.
C//
Most don't NDA's don't outright *forbid* such behavior, but they most certainly prevent it. Look again, carefully, at the details of a typical NDA. Reporting the criminal activity will get you fired under the NDA, and leave you in the position to establish the criminal behavior as a defense against an unlawful firing. Unfortunately, you no longer have a job. Nor are you likely to get one when this shows up on your resume.
... not to mention act as probable cause for an investigation into the company itself.
In many states, it can itself be a crime to not report a felony that you witness. To fire someone for cause for a good faith report on a criminal activity would be highly actionable in every state I'm quite sure,
As for the presence or absence of a network of undermining HR demons in some companies, I won't comment, except to suggest that sharks can swim on both side of that fence. Plenty of blood to be spilled here; no big company can count on mere HR mortals to perjure themselves on the stand, big risk even taking that one to court. If that's indeed what's happening, and they know it, they will settle. Getting caught in perjury and criminal misbehavior in an employment setting will cause a large corporation enormous damages and is total suicide for smaller companies.
C//
haha, what the fuck are they waiting for legal approval for?
Only people in the company who can release the company's intellectual property are its authorized agents. These agents are typically a company executive, who will have fiduciary for the company, and will need to move as carefully as possible. At a software company, one ought to be particularly careful of the company's intellectual property.
The GPL doesn't require that you release your source code; rather it requires that either you do so or face the possible consequences that copyright law implies. There's the larger (I say, larger) issue of community consequences... that's what SWSoft is facing in the immediate. They should at least make a statement of some kind.
C//
For example, a typical NDA prevents you from reporting corporate criminality without a subpoena.
Such an NDA is illegal in every State of the United States.
C//
You're right. You should not have said "most".
Modsim is a unusual use case. That use case has its own concerns.
20% is really on the high end for utilization in virtually every data center.
The syndrome that is most alive today is the "one service, one box" issue. That's why all the drive for consolidation, coming from the virtualization vendors.
C//
Most server farms are running at full speed 24 hours a day.
Afraid not. Data center utilization is typically 20%, and often a lot less. A very lot less.
C//
Or better yet, just line them all up in a single long row, ...
*cough*
I think our data center might be bigger than your data center...
C//
One of my personal points about the 2nd is that it clearly meant, then, what liberals don't want it to mean today. The libertarian interpretation is consistent with the first 150 years of its interpretation. Now, my point: it's fine if the people feel strongly enough to Amend it to mean something less all encompassing today. But the people HAVEN'T DONE THAT. And to have a system of government whereby things you don't like can be stricken due to politics and not procedure means that you really don't have the Constitutional system we were given at all.
C//
ARRRRRG.
You're gonna get me started!
Not only was it only enumerated powers, they felt so strongly about it they passed an Amendment just to make it clear they meant only enumerated powers!
ARGGGGG.
20 Dell 1955 Blades; 16G ram; 70GB SAS 10K drive (one) on which ESX 3.0 is hosted (or a variety of Xen flavors); four gigabit ethernet controllers per blade; CISCO 4948 48 port switch, with 4 ethernets per blade bonded; CISCO 6504e core with Sup-32; Net App 3020 and 3050 for NFS and iSCSI; some EMC Clarion units, likewise.
For CPU we used SPEC CPU 2006 and score about 5-6ish % on VMWare as the same test done on those blades in hard metal. Xen is undiscernably different to the subjective eye than hard metal. I would have to break out large batch testing methodology and run the results through inferential statistics to conclude that there was a difference at all.
I/O is a different story.
The Xen performance claims and the VZ performance claims aren't really useful. They're theoretical. As in, "theoretically, we can stack 100 operating systems on this blade efficiently." Think about that. That's just plain nuts. I can't think of a real use case for that.
BTW, if you like OpenVZ, and have the right use case, the commercial Virtuozzo product ranks as the "best virtualization technology that no one has ever heard of" in my book. They really have their IT management story down pat.
C//