According to a middle-school levelNASA handout about shuttle descent, 16 minutes before landing is the beginning of a maneuver called "roll reversal" to slow down the shuttle. I imagine that this is one of the times that places the highest amount of stress on the shuttle airframe.
CDMA is a general classification of spread-spectrum technologies including PCS and GSM. PCS is crap. I've never been in any area where people are satisfied with their Sprint service. GSM seems to be pretty good, but when I was in England the phone often had an echo problem.
TDMA only tripled the carrying capacity of a cell over analog. What makes digital popular is the error correction. I found that my AT&T phone (TDMA) was dandy for a while but then Chicago became way overcrowded and they didn't keep up on subdividing cells.
If it's under a certain amount (declared by your county) you can seek redress in small claims court. If you want to get the full amount, you might want to ask the VC to participate somehow on your side. Essentially, the VC might be in a position to describe how much money the company should be able to pay you.
Before you go to court, of course, you want to try to settle. And if these guys are holding out on you for much more than the small claims limit, you will want to have your lawyer file a regular suit in a regular manner.
The company may simply not have enough money to pay you and they have a cowardly weasel for a financial officer. If this is probably the case, you should ask what they will pay you, decide if you think it's right. If they are looking for VC you could try to make them sign a promissory note for the remainder, assuming they get funding. This would make them a debtor to you and would change the scene entirely if they actually are going bankrupt.
In any event, you've just discovered the primary reason why consulting should be contracted at high rates. The possibility that you do not get paid is so high you really need insurance against it. So if you went in there thinking, "I made 100k as a salaried programmer, I should be able to make $120k as a contractor," you have probably done yourself a disservice. If you did that, you are likely contracting at $480-$600/day. Contract work, even in these bad times, fetches around $872/day. Good contractors are getting hired somewhere above that number.
Whoa. You incorrectly corrected someone's correct logic. For shame.
N(x) = x is a New Yorker
M(x) = x is an American
1. All x. N(x) -> M(x) [Premise]
2. ~M(a) [Premise]
3. N(a) -> M(a) [Universal Instantiation]
4. ~N(a) QED [Modus Tollens]
So,
I(x) = x has enough intelligence
W(x) = x knows software copying is wrong.
h = this guy.
1. All x. I(x) -> W(x) [Premise]
2. I(h) [Premise]
3. I(h) -> W(h) [Universal Instantiation]
4. W(h) [Modus Ponens]
He knew it was wrong
o = another person
5. flag o [for Universal Generalization]
6. | Assume ~W(o)
7. | | I(o) -> W(o) [Universal Instantiation]
8. | | ~I(o) [Modus Tollens]
9. | ~W(o) -> ~I(o) [Conditional Proof, 6-9]
10.~W(x) -> ~I(x) [Universal Generalization, 5-10]
Anyone who doesn't know it is wrong does not have enough intelligence.
CMU offers, through its Software Engineering Institute, a Masters in Software Engineering. Would I recommend this program? I'm not too sure. The SEI has been developing a document-laden process management methodology for some time and only now are they beginning to see that their way may not be the most effective or may not apply to all situations.
Of course, compared to any other institution, the SEI has done the most research. It's just that they haven't pursued certain directions and therefore you are getting yourself into a very specific way of going about things when you sign onto their way.
The MSE is really like an MBA, I think. You should get it after you've been in the field for some time, when you start showing management potential. Its intention is not to make you into a teriffic software engineer, I think it's to credentialize you as a software engineering manager.
the dpi is going to be the same if the film media is the same but medium format cameras take pictures on 6cm rolls. Most (Hasselblad type) are 6x6 square, some (Fuji type) are 6x9, and there are some 6x4.5 (like the aptly named Maxima 645). 35mm is about 2.5x3.5cm. I don't remember its exact dimensions.
There's a real difference in the effectiveness on your memory of REM sleep as compared to other sleep stages. If you aren't waking up from dreams, you're probably failing to commit things to long-term memory.
Also it is very important to actually do the things you learn in some way or another. For example: people often come in for interviews and they say, "yes, I know java, and design patterns, and all that" but when you quiz them on it they don't have the working knowledge, the know-how. They do know they can look at a book and find it, though. This is slower. Eventually some of those people will develop the working knowledge, and some will do so very quickly. But as an employer, it's better to pay a little more for the safe bet: someone who comes in with that knowledge.
Through counterintelligence, it should be possible to pinpoint potential troublemakers and neutralize them... and neutralize them... and neutralize them... [supposedly Hoover, qtd by RATM]
Of course "and neutralize them before they exercise [realize?] their potential...." is also an interesting extension supposedly to the same line, but the power of that misquoting is nullified a bit by context, "their potential for violence." All of this assuming, of course, it was all said, but the conspiracy theory is still a lot of fun. I think Gates is really saying, "through our intelligence it should be possible to identify competitors' potential patents and neutralize them." But, you know, that's just the rambling of a guy who thinks things are often connected even when they merely appear coincidental.
Now I don't want to get of on a <rant> here, but one can be very handy in reading intentions into another's prose but let us not forget that the simpler way of finding the criminal is to trace the money back from the crime scene. It's a crime to exert your monopoly power to monopolize another market, clear and simple, and it will be a crime until Congress passes a law amending the statute or the courts declare Congress to have been overstepping their boundaries.
Nevertheless, the money was made and a lot of people feel injured. Was it right for them to get burned? Of course it was, provided it was legal. But in this case we have clear evidence of Microsoft previously using their monopoly power to negotiate the monopolization of another market. They were so successful at doing this that they made people think, ten years later, that what they did in this case was right. Never mind that in 1991 Microsoft was already abusing monopoly power over DOS to choke Digital Research (in the irony of ironies way) and the memo clearly states that they were gearing up for a war in which they would launch LAN Manager networking against Novell and DCE, first by initially embracing DCE and then by choking it out. It's a pretty simple plan of attack:
the client for both is included in windows.
You can buy Novell servers, sure, and windows clients, but if you live in that part of the 80s you wouldn't expect much of a network anyway.
Hey, though, we've got this server software coming up that supports DCE and our new networking package, so why don't you hold off on upgrading your Novell, LANtastic, SCO, etc., because you're gonna love this stuff.
Well, of course, when we said, "supports DCE," we meant it supports some aspects of the application-level protocol, but <weasel/>
What do you mean? The fact that networking wasn't part of the operating system was the anomaly. It should have been there.
The fact is, though, that at the time they had a monopoly on pc operating systems without networking. When you have a monopoly you have to play by the rules of monopolies, which means that you can't use it to monopolize other markets. And that starts when you have a monopoly. The punishment for breaking the rules could depend on whether or not you were cognizant of the fact that you were a monopoly, but whether or not you broke the law is only up for legal debate. It doesn't add in any heart-wrenching testimony.
Microsoft's supporters often say, "No, you can't fine Microsoft a lot of money or put heavy constraints on how they do business, break them up or whatever, because they make a lot of money for the economy!" This is irrelevant. They did bad, they should be punished, the punishment should fit the crime and act as a deterrent. Capital gains made by the individuals at the helm of the company during those years (about $50bn, in one case) should be scrutinized and reclaimed. The company should be given a grave decision: split into several competitors or unmonopolize your monopolies.
I think it's very interesting to note that he says that a drop in Microsoft's bucket ($1M) shouldn't be the deciding factor in software spending because the spender may get, God forbid, cheap, poorly built software. Nevertheless, Microsoft make the argument against harsh sentencing and for their actions by saying that people got their requested functionality at a lower price.
The hypocrisy is notable but unimportant. It's OK for a hypocrite to be one in court so long as the positions do not legally contradict one another. No-one ever won an argument by showing the opposition to be hypocritical.
What actually matters is whether we, you know, computer users, have felt that Microsoft's position on what our machines should look like represents not only what we would like to pay for the software but also how it should look. Should it look like Windows XP? Or how about LANtastic 8 running with DR DOS 2002 and Linux clients? Or maybe you would have purchased Macs and Appleshare fileservers running A/UX. Or, perhaps, you would have found the "future" earlier and gone with peer-to-peer networking instead of a client-server model. How much would you have paid for a system that installed as easily as you wanted, worked the way you worked, didn't laugh at you and call you names, etc., etc., etc. Would it have been worth the cost? We don't know because we were precluded from the choice. If people wanted a client-server solution that went for the cost of 1 NT server and five WfW 3.11 clients, I'm sure, and Bill's sure, that someone would have hit that price point. Except Microsoft said, "Well, you need to buy DOS, and no-one else is going to sell DOS, and no-one can build a PC without installing it. We're adding client-server features, gratis, so you're not going to get it cheaper than we'll give it to you. The other company will have to give their product away for free."
At the time, this monopoly (who knew that 95% of the market is a monopoly?) did not have the legal right to make that decision for you. They had to either (1) allow the bundling of DOS with another networking product at the same or lower price or (2) allow someone to sell a PC without DOS. This is what Microsoft should have done. Not because it would have been the nice thing to do, sort of like it'd be nice if a clerk always smiled at a customer, but because it is their legal responsibility, like how it's your neighbor's legal responsibility not to take your newspaper. A nice neighbor doesn't even think about taking your newspaper. But any neighbor who wants to stay within the bounds of the law would not do so either, even if he had a grudge. Novell and DR could have made fun of Microsoft at a party, and made Microsoft really mad, or they could have sold hot dogs where Microsoft expected they'd have a captive hot dog market, but just because you get in their way doesn't mean they should act outside the law until reigned in. Their illegal actions should be punished, for abuses extending back for the past 11 years. And, boy, should the punishment be harsh. </rant>
For the last decade, the government has been purchasing a lot of COTS (Commodity Off-The-Shelf) products, both in the hardware area and in the software area. I'm pretty confident that they are using COTS software for systems that require mandatory (B1+) security. In this case, they're not going to use Windows without heavily modifying the code, but they will use things like Digital Unix that have had mandatory security variants comercially produced. Also, databases like Informix can be delivered with B-level security. The software is commissioned to specification but the specification is often written based on the state of the art. Someone says, "We need a new secure database for this stuff," and they look at a bunch of choices. They decide, perhaps, on Informix, and then the specification looks a lot like Informix.
But, you know, you may disagree. After all, it's easier to justify the tax dollars spent on Defense if you figure that they hand-tool everything to make it just right. I mean, the Pentagon will never actually tell you the real thinking process that goes into buying their software.
Or you project it on a "scope" screen (SP70, I know), and then it comes out just right. I just wish they showed 2001 instead of that dumb roller-coaster film at the Great Escape.
This is, in my opinion, one of the most obnoxious things about the corporate environment. The emergent value of a corporation seems to be to act like a rich person. "Sure, my XJ5 falls apart on the road, but it's a Jag. It's much better than being in a Honda Civic, even if it's souped up like in Fast & the Furious."
There are two reasons to pay a lot for software.
you need it fast
you need it to be perfect
any other reason is simply a problem of an undereducated consumer.
To the poster: if you want to make some good money, try showing up for the bid with finished software. Not complete, but fully functional, and say, "In the meantime, use this." Then, instead of fixed bid, work out an arrangement to get a stream of constant work. With revenue streams, it's much easier to plan partnerships and things to get additional work done.
This model is working for we who distributeQuickFIX. It's a much better model, based on results and trust, rather than on price and appearance.
This is true but I think it'd be very silly of Mandrake, et al, not to court ISVs in partnership. After all, even Microsoft relies on the news from their partners when they are working on features for the OS and its "highly-integrated libraries".
I just love this Libertarian bullshit. It is my firm belief that programmers, by predominantly accepting Libertarian belief systems, have fucked themselves royally. Things like the amendment to the FLSA in 1980 that limited overtime pay to programmers and also exempted most programmers from the protections of the FLSA. In other words, you are treated legally like an executive but you are not paid like one nor are you given such authorities. Many programmers I know are starry-eyed Libertarians who believe that the invisible hand is always at work and that they do not need to worry about the IT industry lobby. Wake the fuck up.
The point is not that Windows is an unstable platform. The point is that a standardized network is at risk of common-mode failure. If all of your ethernet cards are the same, then an unpredicted common failure will affect a percentage of them based on the expected failure rate after discovering the failure. With two varieties, evenly distributed, you will only lose 1/2 of that percentage.
Similarly, if you have standardized on one operating system, an attack that targets that operating system puts 100% of your machines at risk.
An airplane often has redundant parts, not simply duplicated but built by different vendors with different supply chains, etc. This decreases the chance that the secondary device will fail exactly as the primary.
When building a network, one should consider common-mode failures. I don't think it's considered in Gartner's TCO reports.
I find that businesses are often lousy at this. Most business turns a profit when the economy is up and loses money when it is down. A small margin outdoes the economy. That small margin is good at its job, and the rest are mediocre.
So a couple of years back I also wondered about this phenomenon. I've been around the block now, and I've seen managers who hide, managers who pray, managers who try to be buddy-buddy, managers who use Microsoft Project, managers with small and large amounts of tech experience, managers who wage wars, managers who defend territory, managers who protect their "people"....
If you want my guess, I believe that the #1 quality lacking in managers is humility. There is a belief that management responsibility is a reward for good service, and also a tradeoff of making a higher salary. There are few smart companies that will continuously increase a top employee's pay without making him or her into a manager. They are so few that they are almost mythical.
What this has created is a situation where those who express the most interest in becoming managers are the ones who make the most money, and those who want to be managers are usually lesser-qualified employees for managing a creative group than the ones who want to stick around and be creative.
This is a problem, and I don't like it. When I started at my current company, I had a mixture of client and internal management, and the client manager was absolutely the worst. Total climber, no ability, no concern for the team, and quick to blame others when things went wrong. The internal management was not so much better.
Shortly thereafter, I was put on a project where we were asked to fix the problem for our client. I had a nascent set of ideas about the solution, and I got an opportunity to put them into practice.
First off, if you make "manager" mean "administrator" and not "leader", then you empower any member of the team, at any time, to take the leadership role. You also allow any current leader the ability to lay down the role when it no longer suits him or her. The manager can still continue to manage the team, which means trying to get stuff, doing the internal PR campaign, being ultimately responsible for hire/fire/reviews but not the initiator thereof, etc. The manager of a team is in a tough position: no obvious deliverable like an executable program.
Management is also not the same as sponsorship. If there is a stakeholder in the project, that person should play the customer, not the manager. The customer is the one who specifies what is to be done, in what priority. The creative members of the team describe how it is to be done. The manager should not be in between any stakeholder and the team, although it is often a good idea for a small number of stakeholders (i.e. one) to funnel the opinions of all the other stakeholders into a consistent program.
At this point, it should be mentioned that the idea of a "Program Management Office" is generally flawed. If your customer wants to be called that, that's fine, but it doesn't change the fact that he or she should stay on the customer side of the fence.
If you don't have a structure like this, your managers will run in place a lot in order to show that they're sweating. Remember: there is no deliverable, so they get judged on other things. Managers without structure will agree to silly deadlines, exhort the team about performance, tell team members to do stuff, and eventually they will fail. If you're on the outside, it can be fun to watch in a disturbing way, like that scene in "Project X" where they killed the apes.
If you're on the inside, it'll suck like a Hoover.
Can programmers live without management? Yep, we sure can. I do like having them around, though. If a manager appears on your project, send them to meetings in place of the team. Have them informally relate the results. Make sure he or she doesn't make decisions for the team. Make sure you have prepared the manager with the right words to say and a list of things to not discuss. A good manager requires no management, but you often have to handhold ones that are not used to actually doing useful work.
After that, a good manager will be seen and not heard.
So, to answer your question, no, that's not how it should be, but it's how it is 92% of the time. 5% of the time there is a good manager, and 3% there is no manager.
One of the most depressing things about this industry is the typical job solicitation. Here's one, taken verbatim and at random, from HotJobs.com.
DESCRIPTION: Our currency group is looking for a senior programmer with a minimum of 3 years experience developing general user interfaces, calculation routines and data storage and retrieval software. The successful candidate will be a member of a team responsible for the development of the foreign exchange product on the firms system.
REQUIREMENTS: Candidates must have at least 3 years of strong C/Unix experience and 2 years of experience with foreign exchange. Solid communication skills, attention to detail, the ability to work independently as well as within a team, and a willingness to learn are essential. Other programming languages and experience are a plus. BS or higher in Computer Science, Engineering, or Mathematics preferred
This request does nothing to screen out applicants. Of course, it's written rather generally because it is the work of a recruiter, but these sorts of postings are not uncommon anywhere.
In my experience as a mentor, degrees have meant very little. There are a number of universities that are in the business of giving masters' degrees to foreign students so that they can then get H1-B visas. I can't find anything in these degrees in particular that has indicated programming talent. But even more than that, I've often found degree holders to be lost causes when it comes to learning. If they have a degree in software engineering, for example, they tend to believe that they already know how to do it. That is, even if they haven't actually been on a real-life software project yet.
The other thing that they ask for is experience. Experience is fine, but it is also not an indicator of talent. If I've been working in the industry for five years, then I have five years of experience, plain and simple. If I'm a superprogrammer with one year of experience, then I have one year of experience.
As much as it probably pisses people off, one of the best indicators is the psychometrics of problem-solving ability. If you're good at reading comprehension, analytical thinking, problem solving, memory, you're likely to be good at programming. Furthermore, if you have insatiable curiosity for languages, systems, environments, and problem domains, you're likely to continue being talented and increase in skill.
If you think you fit into the above stereotype, try wowing a potential employer with an interesting discussion of what one should actually look for in a programmer. If you like, for $1500 (minus whatever your health insurance pays for) you can get a neuropsychological exam. This is much cheaper than a semester at university, and you might be able to use it to your advantage.
If, however, this exam shows that you're not very good at the basic talents/aptitudes required for programming, you might then consider going to school for a job that suits you better. Otherwise, you'll probably become one of those bitter old programmers no-one likes. -- (MP)
The Feds are the FBI, DOJ, BATF, US Marshalls, etc., who crack down on violations of the law. The Fed is the Federal Reserve. So saying "Fed Raids..." means Alan Greenspan, et al got into the business of fighting software piracy.
Slashdot Mirror
Those who can't, teach! Good for him.
According to a middle-school levelNASA handout about shuttle descent, 16 minutes before landing is the beginning of a maneuver called "roll reversal" to slow down the shuttle. I imagine that this is one of the times that places the highest amount of stress on the shuttle airframe.
TDMA only tripled the carrying capacity of a cell over analog. What makes digital popular is the error correction. I found that my AT&T phone (TDMA) was dandy for a while but then Chicago became way overcrowded and they didn't keep up on subdividing cells.
Excuse me, there's someone knocking on the door... wow, nice suit. Uh oh.
"No, sir, we in the government don't get nice suits."
This guy's dumb.
Before you go to court, of course, you want to try to settle. And if these guys are holding out on you for much more than the small claims limit, you will want to have your lawyer file a regular suit in a regular manner.
The company may simply not have enough money to pay you and they have a cowardly weasel for a financial officer. If this is probably the case, you should ask what they will pay you, decide if you think it's right. If they are looking for VC you could try to make them sign a promissory note for the remainder, assuming they get funding. This would make them a debtor to you and would change the scene entirely if they actually are going bankrupt.
In any event, you've just discovered the primary reason why consulting should be contracted at high rates. The possibility that you do not get paid is so high you really need insurance against it. So if you went in there thinking, "I made 100k as a salaried programmer, I should be able to make $120k as a contractor," you have probably done yourself a disservice. If you did that, you are likely contracting at $480-$600/day. Contract work, even in these bad times, fetches around $872/day. Good contractors are getting hired somewhere above that number.
Good luck!
Whoa. You incorrectly corrected someone's correct logic. For shame. N(x) = x is a New Yorker
M(x) = x is an American
1. All x. N(x) -> M(x) [Premise]
2. ~M(a) [Premise]
3. N(a) -> M(a) [Universal Instantiation]
4. ~N(a) QED [Modus Tollens]
So,
I(x) = x has enough intelligence
W(x) = x knows software copying is wrong.
h = this guy.
1. All x. I(x) -> W(x) [Premise]
2. I(h) [Premise]
3. I(h) -> W(h) [Universal Instantiation]
4. W(h) [Modus Ponens]
He knew it was wrong
o = another person
5. flag o [for Universal Generalization]
6. | Assume ~W(o)
7. | | I(o) -> W(o) [Universal Instantiation]
8. | | ~I(o) [Modus Tollens]
9. | ~W(o) -> ~I(o) [Conditional Proof, 6-9]
10.~W(x) -> ~I(x) [Universal Generalization, 5-10]
Anyone who doesn't know it is wrong does not have enough intelligence.
Whoa. You incorrectly corrected someone's correct logic. For shame. N(x) = x is a New Yorker M(x) = x is an American 1. All x. N(x) -> M(x) [Premise] 2. ~M(a) [Premise] 3. N(a) -> M(a) [Universal Instantiation] 4. ~N(a) QED [Modus Tollens] So, I(x) = x has enough intelligence W(x) = x knows software copying is wrong. h = this guy. 1. All x. I(x) -> W(x) [Premise] 2. I(h) [Premise] 3. I(h) -> W(h) [Universal Instantiation] 4. W(h) [Modus Ponens] He knew it was wrong o = another person 5. flag o [for Universal Generalization] 6. | Assume ~W(o) 7. | | I(o) -> W(o) [Universal Instantiation] 8. | | ~I(o) [Modus Tollens] 9. | ~W(o) -> ~I(o) [Conditional Proof, 6-9] 10.~W(x) -> ~I(x) [Universal Generalization, 5-10] Anyone who doesn't know it is wrong does not have enough intelligence.
Of course, compared to any other institution, the SEI has done the most research. It's just that they haven't pursued certain directions and therefore you are getting yourself into a very specific way of going about things when you sign onto their way.
The MSE is really like an MBA, I think. You should get it after you've been in the field for some time, when you start showing management potential. Its intention is not to make you into a teriffic software engineer, I think it's to credentialize you as a software engineering manager.
the dpi is going to be the same if the film media is the same but medium format cameras take pictures on 6cm rolls. Most (Hasselblad type) are 6x6 square, some (Fuji type) are 6x9, and there are some 6x4.5 (like the aptly named Maxima 645). 35mm is about 2.5x3.5cm. I don't remember its exact dimensions.
There's a real difference in the effectiveness on your memory of REM sleep as compared to other sleep stages. If you aren't waking up from dreams, you're probably failing to commit things to long-term memory.
Also it is very important to actually do the things you learn in some way or another. For example: people often come in for interviews and they say, "yes, I know java, and design patterns, and all that" but when you quiz them on it they don't have the working knowledge, the know-how. They do know they can look at a book and find it, though. This is slower. Eventually some of those people will develop the working knowledge, and some will do so very quickly. But as an employer, it's better to pay a little more for the safe bet: someone who comes in with that knowledge.
[supposedly Hoover, qtd by RATM]
Of course "and neutralize them before they exercise [realize?] their potential...." is also an interesting extension supposedly to the same line, but the power of that misquoting is nullified a bit by context, "their potential for violence." All of this assuming, of course, it was all said, but the conspiracy theory is still a lot of fun. I think Gates is really saying, "through our intelligence it should be possible to identify competitors' potential patents and neutralize them." But, you know, that's just the rambling of a guy who thinks things are often connected even when they merely appear coincidental.
Now I don't want to get of on a <rant> here, but one can be very handy in reading intentions into another's prose but let us not forget that the simpler way of finding the criminal is to trace the money back from the crime scene. It's a crime to exert your monopoly power to monopolize another market, clear and simple, and it will be a crime until Congress passes a law amending the statute or the courts declare Congress to have been overstepping their boundaries.
Nevertheless, the money was made and a lot of people feel injured. Was it right for them to get burned? Of course it was, provided it was legal. But in this case we have clear evidence of Microsoft previously using their monopoly power to negotiate the monopolization of another market. They were so successful at doing this that they made people think, ten years later, that what they did in this case was right. Never mind that in 1991 Microsoft was already abusing monopoly power over DOS to choke Digital Research (in the irony of ironies way) and the memo clearly states that they were gearing up for a war in which they would launch LAN Manager networking against Novell and DCE, first by initially embracing DCE and then by choking it out. It's a pretty simple plan of attack:
- the client for both is included in windows.
- You can buy Novell servers, sure, and windows clients, but if you live in that part of the 80s you wouldn't expect much of a network anyway.
- Hey, though, we've got this server software coming up that supports DCE and our new networking package, so why don't you hold off on upgrading your Novell, LANtastic, SCO, etc., because you're gonna love this stuff.
- Well, of course, when we said, "supports DCE," we meant it supports some aspects of the application-level protocol, but <weasel/>
- What do you mean? The fact that networking wasn't part of the operating system was the anomaly. It should have been there.
The fact is, though, that at the time they had a monopoly on pc operating systems without networking. When you have a monopoly you have to play by the rules of monopolies, which means that you can't use it to monopolize other markets. And that starts when you have a monopoly. The punishment for breaking the rules could depend on whether or not you were cognizant of the fact that you were a monopoly, but whether or not you broke the law is only up for legal debate. It doesn't add in any heart-wrenching testimony.Microsoft's supporters often say, "No, you can't fine Microsoft a lot of money or put heavy constraints on how they do business, break them up or whatever, because they make a lot of money for the economy!" This is irrelevant. They did bad, they should be punished, the punishment should fit the crime and act as a deterrent. Capital gains made by the individuals at the helm of the company during those years (about $50bn, in one case) should be scrutinized and reclaimed. The company should be given a grave decision: split into several competitors or unmonopolize your monopolies.
I think it's very interesting to note that he says that a drop in Microsoft's bucket ($1M) shouldn't be the deciding factor in software spending because the spender may get, God forbid, cheap, poorly built software. Nevertheless, Microsoft make the argument against harsh sentencing and for their actions by saying that people got their requested functionality at a lower price.
The hypocrisy is notable but unimportant. It's OK for a hypocrite to be one in court so long as the positions do not legally contradict one another. No-one ever won an argument by showing the opposition to be hypocritical.
What actually matters is whether we, you know, computer users, have felt that Microsoft's position on what our machines should look like represents not only what we would like to pay for the software but also how it should look. Should it look like Windows XP? Or how about LANtastic 8 running with DR DOS 2002 and Linux clients? Or maybe you would have purchased Macs and Appleshare fileservers running A/UX. Or, perhaps, you would have found the "future" earlier and gone with peer-to-peer networking instead of a client-server model. How much would you have paid for a system that installed as easily as you wanted, worked the way you worked, didn't laugh at you and call you names, etc., etc., etc. Would it have been worth the cost? We don't know because we were precluded from the choice. If people wanted a client-server solution that went for the cost of 1 NT server and five WfW 3.11 clients, I'm sure, and Bill's sure, that someone would have hit that price point. Except Microsoft said, "Well, you need to buy DOS, and no-one else is going to sell DOS, and no-one can build a PC without installing it. We're adding client-server features, gratis, so you're not going to get it cheaper than we'll give it to you. The other company will have to give their product away for free."
At the time, this monopoly (who knew that 95% of the market is a monopoly?) did not have the legal right to make that decision for you. They had to either (1) allow the bundling of DOS with another networking product at the same or lower price or (2) allow someone to sell a PC without DOS. This is what Microsoft should have done. Not because it would have been the nice thing to do, sort of like it'd be nice if a clerk always smiled at a customer, but because it is their legal responsibility, like how it's your neighbor's legal responsibility not to take your newspaper. A nice neighbor doesn't even think about taking your newspaper. But any neighbor who wants to stay within the bounds of the law would not do so either, even if he had a grudge. Novell and DR could have made fun of Microsoft at a party, and made Microsoft really mad, or they could have sold hot dogs where Microsoft expected they'd have a captive hot dog market, but just because you get in their way doesn't mean they should act outside the law until reigned in. Their illegal actions should be punished, for abuses extending back for the past 11 years. And, boy, should the punishment be harsh. </rant>
For the last decade, the government has been purchasing a lot of COTS (Commodity Off-The-Shelf) products, both in the hardware area and in the software area. I'm pretty confident that they are using COTS software for systems that require mandatory (B1+) security. In this case, they're not going to use Windows without heavily modifying the code, but they will use things like Digital Unix that have had mandatory security variants comercially produced. Also, databases like Informix can be delivered with B-level security. The software is commissioned to specification but the specification is often written based on the state of the art. Someone says, "We need a new secure database for this stuff," and they look at a bunch of choices. They decide, perhaps, on Informix, and then the specification looks a lot like Informix.
But, you know, you may disagree. After all, it's easier to justify the tax dollars spent on Defense if you figure that they hand-tool everything to make it just right. I mean, the Pentagon will never actually tell you the real thinking process that goes into buying their software.
Or you project it on a "scope" screen (SP70, I know), and then it comes out just right. I just wish they showed 2001 instead of that dumb roller-coaster film at the Great Escape.
What about the current crop of software running nuclear weapons stations and reactors make you feel most secure?
but why is this a troll?
There are two reasons to pay a lot for software.
- you need it fast
- you need it to be perfect
any other reason is simply a problem of an undereducated consumer.To the poster: if you want to make some good money, try showing up for the bid with finished software. Not complete, but fully functional, and say, "In the meantime, use this." Then, instead of fixed bid, work out an arrangement to get a stream of constant work. With revenue streams, it's much easier to plan partnerships and things to get additional work done.
This model is working for we who distributeQuickFIX. It's a much better model, based on results and trust, rather than on price and appearance.
This is true but I think it'd be very silly of Mandrake, et al, not to court ISVs in partnership. After all, even Microsoft relies on the news from their partners when they are working on features for the OS and its "highly-integrated libraries".
I just love this Libertarian bullshit. It is my firm belief that programmers, by predominantly accepting Libertarian belief systems, have fucked themselves royally. Things like the amendment to the FLSA in 1980 that limited overtime pay to programmers and also exempted most programmers from the protections of the FLSA. In other words, you are treated legally like an executive but you are not paid like one nor are you given such authorities. Many programmers I know are starry-eyed Libertarians who believe that the invisible hand is always at work and that they do not need to worry about the IT industry lobby. Wake the fuck up.
The point is not that Windows is an unstable platform. The point is that a standardized network is at risk of common-mode failure. If all of your ethernet cards are the same, then an unpredicted common failure will affect a percentage of them based on the expected failure rate after discovering the failure. With two varieties, evenly distributed, you will only lose 1/2 of that percentage.
Similarly, if you have standardized on one operating system, an attack that targets that operating system puts 100% of your machines at risk.
An airplane often has redundant parts, not simply duplicated but built by different vendors with different supply chains, etc. This decreases the chance that the secondary device will fail exactly as the primary.
When building a network, one should consider common-mode failures. I don't think it's considered in Gartner's TCO reports.
I find that businesses are often lousy at this. Most business turns a profit when the economy is up and loses money when it is down. A small margin outdoes the economy. That small margin is good at its job, and the rest are mediocre.
Oooh, i like it like that
So a couple of years back I also wondered about this phenomenon. I've been around the block now, and I've seen managers who hide, managers who pray, managers who try to be buddy-buddy, managers who use Microsoft Project, managers with small and large amounts of tech experience, managers who wage wars, managers who defend territory, managers who protect their "people"....
If you want my guess, I believe that the #1 quality lacking in managers is humility. There is a belief that management responsibility is a reward for good service, and also a tradeoff of making a higher salary. There are few smart companies that will continuously increase a top employee's pay without making him or her into a manager. They are so few that they are almost mythical.
What this has created is a situation where those who express the most interest in becoming managers are the ones who make the most money, and those who want to be managers are usually lesser-qualified employees for managing a creative group than the ones who want to stick around and be creative.
This is a problem, and I don't like it. When I started at my current company, I had a mixture of client and internal management, and the client manager was absolutely the worst. Total climber, no ability, no concern for the team, and quick to blame others when things went wrong. The internal management was not so much better.
Shortly thereafter, I was put on a project where we were asked to fix the problem for our client. I had a nascent set of ideas about the solution, and I got an opportunity to put them into practice.
First off, if you make "manager" mean "administrator" and not "leader", then you empower any member of the team, at any time, to take the leadership role. You also allow any current leader the ability to lay down the role when it no longer suits him or her. The manager can still continue to manage the team, which means trying to get stuff, doing the internal PR campaign, being ultimately responsible for hire/fire/reviews but not the initiator thereof, etc. The manager of a team is in a tough position: no obvious deliverable like an executable program.
Management is also not the same as sponsorship. If there is a stakeholder in the project, that person should play the customer, not the manager. The customer is the one who specifies what is to be done, in what priority. The creative members of the team describe how it is to be done. The manager should not be in between any stakeholder and the team, although it is often a good idea for a small number of stakeholders (i.e. one) to funnel the opinions of all the other stakeholders into a consistent program.
At this point, it should be mentioned that the idea of a "Program Management Office" is generally flawed. If your customer wants to be called that, that's fine, but it doesn't change the fact that he or she should stay on the customer side of the fence.
If you don't have a structure like this, your managers will run in place a lot in order to show that they're sweating. Remember: there is no deliverable, so they get judged on other things. Managers without structure will agree to silly deadlines, exhort the team about performance, tell team members to do stuff, and eventually they will fail. If you're on the outside, it can be fun to watch in a disturbing way, like that scene in "Project X" where they killed the apes.
If you're on the inside, it'll suck like a Hoover.
Can programmers live without management? Yep, we sure can. I do like having them around, though. If a manager appears on your project, send them to meetings in place of the team. Have them informally relate the results. Make sure he or she doesn't make decisions for the team. Make sure you have prepared the manager with the right words to say and a list of things to not discuss. A good manager requires no management, but you often have to handhold ones that are not used to actually doing useful work.
After that, a good manager will be seen and not heard.
So, to answer your question, no, that's not how it should be, but it's how it is 92% of the time. 5% of the time there is a good manager, and 3% there is no manager.
Cheers,
John
This request does nothing to screen out applicants. Of course, it's written rather generally because it is the work of a recruiter, but these sorts of postings are not uncommon anywhere.
In my experience as a mentor, degrees have meant very little. There are a number of universities that are in the business of giving masters' degrees to foreign students so that they can then get H1-B visas. I can't find anything in these degrees in particular that has indicated programming talent. But even more than that, I've often found degree holders to be lost causes when it comes to learning. If they have a degree in software engineering, for example, they tend to believe that they already know how to do it. That is, even if they haven't actually been on a real-life software project yet.
The other thing that they ask for is experience. Experience is fine, but it is also not an indicator of talent. If I've been working in the industry for five years, then I have five years of experience, plain and simple. If I'm a superprogrammer with one year of experience, then I have one year of experience.
As much as it probably pisses people off, one of the best indicators is the psychometrics of problem-solving ability. If you're good at reading comprehension, analytical thinking, problem solving, memory, you're likely to be good at programming. Furthermore, if you have insatiable curiosity for languages, systems, environments, and problem domains, you're likely to continue being talented and increase in skill.
If you think you fit into the above stereotype, try wowing a potential employer with an interesting discussion of what one should actually look for in a programmer. If you like, for $1500 (minus whatever your health insurance pays for) you can get a neuropsychological exam. This is much cheaper than a semester at university, and you might be able to use it to your advantage.
If, however, this exam shows that you're not very good at the basic talents/aptitudes required for programming, you might then consider going to school for a job that suits you better. Otherwise, you'll probably become one of those bitter old programmers no-one likes. -- (MP)
The Feds are the FBI, DOJ, BATF, US Marshalls, etc., who crack down on violations of the law. The Fed is the Federal Reserve. So saying "Fed Raids..." means Alan Greenspan, et al got into the business of fighting software piracy.