Re:.Net Runtime negates the need for this
on
Microsoft's New Hurdles
·
· Score: 3, Informative
Microsoft will probably start giving away a *nix-based.Net runtime first. Once you have all your products running on an abstraction layer, the OS becsome irrelevant.
Like this one you mean? Microsoft have been giving away the a.NET runtime and development environment (including source) for FreeBSD since last March.
Look at their product line. Their base-level SunBlade 2000 workstation has a 1GHz processor, 1GB of memory, a 73GB disk drive, a good graphics board, comes with StarOffice, and costs $10,995. Keyboard and monitor extra. What's wrong with this picture?
You can get a Sun Blade 100 for about a thousand dollars . That's quite competitive. You can't compare these Mhz-for-Mhz with PCs remember.
Instead, Sun needs to go after.NET with a Java3 marketing blitz, before.NET gets too established. Take all that money they're spending on lawyers and saturate the enterprise app market with advertising and FUD.
Umm, they are, with Sun ONE, have a click around their web site. Unfortunately, they're taking competing with.NET too literally and there's little strategic focus. As a company that (still) makes most of its money on hardware, they should be pushing - hard - why it makes sense to have Sun Blades and Rays on desktops and Sun Fires in the data center, all integrated with ONE, but their message is fragmented and confusing at best.
They might also need to stop competing with everyone. Maybe make IRIX one of their supported platforms - easier than Win32, which they've had a JDK for since day one. But not Tru64 or HPUX since Compaq is a company they actually do compete directly with - hardware + OS + application software. The perfect integration of OS with hardware is something they have that MS don't. Maybe try to repair their relationship with Oracle, and start courting the workstation software vendors who are busily porting their formerly Unix-only apps to Windows (altho' that will probably require getting the SPARC's FP performance well above the x86).
Let me clarify that I'm absolutely not joking - I'm a J2EE consultant, and I really like the technology; I don't savor the prospect of having to become a.NET consultant to pay the bills in 3 or 4 years.
I fully expect a third party to create a Java to CLR compiler at some point, it should be quite straightforward to do, the only thing I can think of that's prevented it is the fear of lawsuits. In any case, I think you're safe in the 3-4 year horizon, Java is the new COBOL i.e. the de facto corporate development language.
Netscape clearly had the better product. It was fast, small, and had the latest bells and whistles like Java, Javascript, cgi support, and a whole host of other things we take for granted today.
It's tempting to blame Microsoft for Netscape's failure, but it's not what happened. To understand this, you must understand Netscape's strategy: to create demand for server products. What mattered to the business plan was there were lots of people with browsers demanding content from servers, and Netscape developed their browser for the sole reason of seeding the market.
I'm not making this up, here are some quotes from the founders:
Skeptics continue to wonder about Netscape's strategy: How can a company that aspires to dominate a market give away its core product? In fact the Navigator, while certainly Netscape's most famous product, is not its core product. The Navigator is the market-maker by which Netscape establishes a standard. Its growing collection of server products -- complex software that companies use to post information on the Web and conduct electronic commerce --are the revenue generators through which Netscape will earn the bulk of its profits.
"Netscape builds printing presses," says President James Barksdale. "But first I've got to teach everybody to read, or there won't be any publishers."
Jim Clark offers a simpler explanation. "This is not freeware," he says, "this is marketware."
What killed Netscape is that they overreached themselves; version 1.x of their server products were the first serious products on the market, version 2.x were very good (NES 2.01 is still probably the best webserver I have deployed) but 3.x were terrible: buggy and slow. That's what caused their collapse, people stopped buying their server products. By the time they lost the browser war, it was irrelevant anyway.
Whatever you might feel about McNealy personally, ya gotta give him credit for sheer guts and having razor-sharp focus IMHO
I guess you haven't been following Sun for very long, because McNealy and "focus" don't get along. Sun has flipflopped for years unable to decide if they were a workstation vendor, a server vendor, a vertical one-stop shop a la IBM or something else altogether (see their current marketing campaign around SunONE). The "dot in dotcom" was a disaster for them, and now their sales are being hurt by their own second-hand kit (why buy a brand new Sun Blade workstation when you can pick up a very capable Ultra 60 or 30 for a fraction of the price?). They spent years insisting that Solaris on SPARC was their crown jewels, they fudged the decision on Solaris x86 and now they are selling Linux on x86 machines, this time competing with themselves directly. Or StarOffice - that was part of a plan to compete directly with PCs using cheap workstations running Solaris (the Ultra 5) and thin clients (Sun Ray). It didn't pan out, and they were pretty much forced to give away StarOffice for free (you think they did that out of the goodness of their hearts?). Over the past couple of years their development toolset has been called SPARCworks, SunPro, Forte and now it's SunONE Studio - the same basic tools rebranded depending on whether Sun think the future is embedded, HPC, Internet or "ONE". And Java - for a long time, IBM's JVM on x86 wiped the floor with Sun's on SPARC (may even still be true, haven't checked recently).
Sun have some great products and technologies, but what will kill them is their sheer lack of focus. Right now they need to reestablish the SPARC as the processor for serious computing and establish the Sun Fire/Blade as the platforms for it, and they need to get Java's performance on their low end machines up to the level where it can seriously be used for interactive apps then start selling them in bulk to the people who are currently buying their workstations from Dell.
But the very first thing they need if for McNealy to drop his personal feud with Microsoft and define his company in terms of what his customers want, not simply reacting to whatever MS do.
Same thing with mouse tracking - the X display refresh is not synced in any way with mouse position updates, so moving a window around feels "sticky."
This is actually fairly fundamental to the X paradigm, which is network-oriented. There needs to be no logic on the display device, so things like mouse movements are transmitted over the network for the remote application to see and deal with. It doesn't matter to X if the network is TCP/IP over a WAN or your localhost loopback adaptor. Therefore, it has to be asynchronous, or X would appear to just "freeze" while it was waiting for the network - better that it gives you some feedback and lets you keep working while it sorts itself out than to make you wait for proper synchonization of all GUI updates.
Sun solved this with NeWS which actually does put some interactivity on the display device - for example, you could click a button and the animation of it being "pressed" would be processed on the display and just the button-pressed event sent back over the network (X will have to send mouse-moved and button-pressed events one way, and drawing instructions to display the button pressing over the network). But NeWS never caught on because other workstation vendors weren't prepared to let Sun control the standard.
X is powerful, but it was never made to be fast, and never will be without sacrificing some flexibility. From the Windows perspective the opposite is true: it was made to be fast, and can't be made as flexible as X without sacrificing speed (try using an XP remote desktop over a WAN, for example, or PC Anywhere).
The lady in the booth paused and said, "That's Duane!" She said she recognized my walk.
This is why US special forces can't infiltrate Afghanistan in burkas - they don't walk like women. The British army had a similar problem in Northern Ireland - undercover soldiers tended to march rather than walk, and stood straighter than the laid-back locals, so stood out. So there are definitely applications for this technology, but unfortunately it seems like it would be more useful to terrorists than our side.
But they are not a "public good", they are something that only drivers want and need, yet everybody effectively has to pay for roads and other driving related costs.
How amusing. I take it you've never bought anything from a store? How do you suppose it got there, hmmm?
Rail works if you have:
large amounts of stuff
that you need to move in relatively straight lines
between predefined points
on a regular schedule
without great urgency
Fortunately, lots of people do have such stuff, or even are such stuff themselves, but over short distances between many points with arbitrary journey start times and routes, mass transit systems are the wrong paradigm.
Sadly, a lot of areas of the US are built in a way that you can't do without a car anymore. I cannot afford to live an area where I can walk to the post office or to a store. But that's not an "advantage".
Sure you can - but you can't expect the benefits of big-city living in a small community. There are plenty of small towns in the US, where people say hello to each other in the streets and the waitress in the diner remembers your name.
But you cannot expect the world to adapt itself to your personal preferences. Either you live in a sprawling metropolis and have access to all it offers, or you live in a small town and adopt the lifestyle. Either way, you make your own choices.
The answer is pretty simple - bad management. Most points that you made (and which I consider completely valid, especially given your impressive experience), can be in fact reduced to bad management. Good management is the exception, not the rule.
I see this asserted time and time again, but experience leads me to the conclusion that if it just as often incompetent workers as incompetent managers that cause projects and hence companies to fail.
For example, management sometimes makes unrealistic claims to customers about what the product can do and when it will be ready, but no more unreasonable or unrealistic than the claims made by programmers to project managers that a particular feature will be ready by a certain date, or even that it will work as it should. Some managers abuse their position and authority by indulging themselves using corporate resources, but no more that prima donna programmers who demand exemptions from the company dress code, free Snapple, aromatherapy etc. Sometimes management demands that the programming staff works long hours, but no more often than the programmers decide to blow off work for the day and play games on the LAN, surf Slashdot, chat on IRC etc.
The true exception is the company where both management and workers have dropped the juvenile "them and us" attitude and work in harmony. There are more than a few of them out there. Usually they are created by the founders not for money per se, but for their own freedom, and are at least their third company. In such a company both management and workers recognise that they have a good thing going here and that it is up to everyone involved to keep it that way.
I know the lines between the two have begun to blur, now that desktop CPU's have vector instructions (SSE/3DNow); is there something left that desktop PC's (even clusters of them) just don't offer?
You can only horizontally scale a stateless application. For example, you could use a different node to render each scene in a movie - but it would be very difficult to use one machine to render trees throughout the entire movie and another to compute their shadows, because of the interaction between them. Trees create shadows, and shadows fall across trees.
Stateful applications require constant synchronization between computational elements, and for that, you need to vertically scale in something with a nice fast active backplane, otherwise your app will spend more time shuffling data around than working.
I'm looking for a (cheap) solution for filesystem sharing between two linux servers and, since the target is just redundancy, I've come to the following idea
Before you spend a single dollar, ask yourself: if your system is important enough to require fault tolerance, why can't you spend money to get a professional solution? If your system isn't important enough to spend money on, then ordinary bidirectional file replication should be good enough for you. You could do it with rsync and ntpd in a few minutes, for free.
Ever think it might be sort of cool to run that kind of low-fat system on powerful modern hardware? Sort of like an ultimate C64 with, hundreds of megabytes of RAM, and many hundreds of MIPS. Maybe not pretty, but the raw computing horsepower at your fingertips would be awesome, provided you know the system well enough to use it
Couldn't you do this by booting a modern PC into DOS? Running compiled C it would be blazingly fast, your program would get literally 100% of the machine's capacity.
I've seen the acronym "TA" appear in a few articles recently - could someone please explain it for me. Ty
Teaching Assistant. It's common for tenured academic faculty to get their grad students to do the bread-and-butter teaching work, like marking exam papers. TAs are the bottom of the academic pile. It's little better than indentured servitude, but they have to suck it up or get on the bad side of the prof who holds their fate in his hands. TAs are miserable, but everyone's gotta pay their dues.
Unions, implemented correctly, start and end with that sentiment. This "rugged individualism" (rugged geekdom?) plays well on TV, but doesn't scale to real life. We've all seen that typical geek skills are becoming more common and less valuable
Answer me this: if your skills are less valuable, by what logic should you even be paid the same for them, let alone more?
There's actually little reason for your skills portfolio to become less valuable over time. Let me give you example. When I started in IT (1996), HTML and Perl CGI were in-demand skills. Very few people had them, lots of people wanted them, those that did have them were well paid for them by those who wanted them. Nowadays, of course, it's difficult to find someone who doesn't have HTML, and Perl CGI isn't really in demand anymore now that PHP, Cold Fusion, JSP et al are used.
But, I'm not unemployed, because I kept my skills up to date (FWIW, I do mission-critical OLTP work for investment banks). Someday, that'll be obsolete too - but I'm not even worried slightly, in 5 years I'll have a whole new skill set, and I'll still be ahead of the curve.
What would a union do? Demand, backed by threats of strikes and pickets, that companies should stick with old-fashioned, labor-intensive technologies to save their members from having to develop their skills? Demand that only union-members are employed, then restrict membership to drive up the price, even though there are people with the same skills happy to do the work for less? That's what happens in unionized industries. It rests on the assumption that all workers are equal and are entitled to equal treatment.
But that falls apart in any business where individuals can have an impact on productivity, and those individuals know that by freely associating, they can get a lot more done. You can see this time and time again - mini-mills outproducing Big Steel is a typical case. The buyers will go to where they can get the most for the least - that's rational economic behavior. Without government protection, customers will leave inefficient union shops in droves. Without customers but the union insisting that the payroll isn't cut, the organization will lose money, eating into its reserves. Eventually those will be gone, and rather than a few people losing their jobs, everyone will lose their jobs. If there are no subsidies forthcoming from the taxpayer, the employer will collapse.
RIAA is the antithesis of free market. RIAA is an organization of media companies who band together to fix prices and shutout competition. RIAA is why you cannot find music published by smaller record labels in music shops.
The RIAA is a cartel, just like De Beers. A cartel is what it's called when sellers in a marketplace agree amongst themselves not to compete with each other on price and often on quality, thereby inflating the price and reducing quality to the lowest common denominator. This is because it is not in the economic interest of any member of the cartel to provide a better quality than any other member, since they won't be rewarded for doing so. A marketplace should act in the best interests of both sellers and buyers, but a cartel skews the market against the buyers. In most regulated markets, cartels are illegal.
Hiring labor is a market like any other. There is a buyer (called an "employer"), and a seller (an "employee"). The seller is selling time, and the buyer is buying it.
The only difference between a cartel and a union is that unions are legal. It is no coincidence that the heavily unionized industries - auto manufacturing, steel, coal mining - have been wrecked, because the internal market that rewards quality and implicity punishes poor quality has been destroyed by the activity of the cartel. A company is an organization for generating economic value. If productivity is not the criteria by which it is measured, instead it employees people to artificially swell the payroll even though they do not produce more than they cost, the cost/benefit calculation fails, and its resources cannot be allocated efficiently. It is doomed therefore to burn its capital, and will implode when that is exhausted. Like any protectionist policy, a union can only protect the interests of its members in the short term, because it is inherently less efficient that a free market in labor.
No - what will happen is that corporations will incorporate trade barriers into the product (such as DVD region coding).
Well, they certainly tried to, but the market responded by creating then satisfying a demand for multi-region players. Without government backing, you can't prevent arbitrage.
given that despite the fact that it's glaringly obvious that UK Brits are systematically ripped off on everything from Cars to Computers compared to our European counterparts - very little action is taken.
Well, duh, who do you suppose is doing the ripping off? Of every pound you spend on petrol, 80p is tax - that's a 400% tax! A pack of cigarettes costs something like 25p to manufacture - and you're paying 15x that in tax!
And everyone pays, even non-drivers - because everything is shipped by road, it just gets factored into the price of loaves of bread. And, despite what the propaganda tells you, the NHS would be well and truly shafted without money from cigarette taxes - the government really wants more smokers, not less. After all they are the perfect citizens: they voluntarily pay more tax their whole lives, then die before they collect their pensions!
Parliament will take no action on rip-off Britain because they are it's biggest fans! And that's why in the past Customs have been so heavy handed with "booze cruisers" - it's only because of recent public outrage and the impossbility of comprehensive enforcement that the rules have recently been (slightly) relaxed.
With that said, I swear to god, multinational cooperations have no conscience. Turn on the news, and all you see is the Enrons, Microsofts, and all these other coopertions who do everything they can to screw the consumer and their employees to make an extra penny. Good for the Europeans, bout damn time someone smacked those companies down, even if it is one with good Karma like nintendo.
On the contrary, multinationals are only operating within the framework provided by national governments. When governments dismantle their trade barriers, such as import tarriffs and quotas, then price differences will simply be arbitraged away by brokers (i.e. you see something selling for $10 in country A and $5 in country B, export/import it and sell it in country A for $6 - eventually the margin will tend to zero). But that can only happen if there are no obstacles to freely moving goods and capital around.
The biggest barrier to this is ironically the EU itself who protect manufacturers like Levi Strauss from UK retailers who source overseas and want to sell at less than Levi's MRRP. Not to mention the distortions the EU create in the market with their subsidies of inefficient industries.
Frankly, I don't know who's worse, corrupt corporations (as distinct from well-run corporations) or corrupt politicians - and the EU isn't even democratically elected! A shareholder has far more influence on a company than a voter has on the European Commission (that's a fact).
Microsoft once (still does?) own a stake of SCO. I remember seeing a copyright Microsoft somewhere either in the OS or in the documentation once.
IIRC, Microsoft sold Xenix to SCO. Xenix was Microsoft Unix running on the 286. Microsoft agreed with AT&T (or someone) not to develop their own Unix.
Microsoft, in all it's *nix hating glory owned a portion of a *nix company
Microsoft don't hate Unix per se; a lot of early MS development (again IIRC) was done on Unix-running DECs cross-compiling to 8086. It's just not their own product, and therefore they have a pathological urge to compete with it.
The big disadvantage to large laser weapons is that they give away their precise position since laser beams travel in perfectly straight lines.
You might think so, but tracer ammo has been around since WW2 and it's still in use today. That suggests that being able to locate a weapon that is firing at you while it is firing isn't as big a tactical advantage as it might first appear.
Slashdot Mirror
Microsoft will probably start giving away a *nix-based .Net runtime first. Once you have all your products running on an abstraction layer, the OS becsome irrelevant.
.NET runtime and development environment (including source) for FreeBSD since last March.
Like this one you mean? Microsoft have been giving away the a
Look at their product line. Their base-level SunBlade 2000 workstation has a 1GHz processor, 1GB of memory, a 73GB disk drive, a good graphics board, comes with StarOffice, and costs $10,995. Keyboard and monitor extra. What's wrong with this picture?
You can get a Sun Blade 100 for about a thousand dollars . That's quite competitive. You can't compare these Mhz-for-Mhz with PCs remember.
Instead, Sun needs to go after .NET with a Java3 marketing blitz, before .NET gets too established. Take all that money they're spending on lawyers and saturate the enterprise app market with advertising and FUD.
.NET too literally and there's little strategic focus. As a company that (still) makes most of its money on hardware, they should be pushing - hard - why it makes sense to have Sun Blades and Rays on desktops and Sun Fires in the data center, all integrated with ONE, but their message is fragmented and confusing at best.
.NET consultant to pay the bills in 3 or 4 years.
Umm, they are, with Sun ONE, have a click around their web site. Unfortunately, they're taking competing with
They might also need to stop competing with everyone. Maybe make IRIX one of their supported platforms - easier than Win32, which they've had a JDK for since day one. But not Tru64 or HPUX since Compaq is a company they actually do compete directly with - hardware + OS + application software. The perfect integration of OS with hardware is something they have that MS don't. Maybe try to repair their relationship with Oracle, and start courting the workstation software vendors who are busily porting their formerly Unix-only apps to Windows (altho' that will probably require getting the SPARC's FP performance well above the x86).
Let me clarify that I'm absolutely not joking - I'm a J2EE consultant, and I really like the technology; I don't savor the prospect of having to become a
I fully expect a third party to create a Java to CLR compiler at some point, it should be quite straightforward to do, the only thing I can think of that's prevented it is the fear of lawsuits. In any case, I think you're safe in the 3-4 year horizon, Java is the new COBOL i.e. the de facto corporate development language.
It's tempting to blame Microsoft for Netscape's failure, but it's not what happened. To understand this, you must understand Netscape's strategy: to create demand for server products. What mattered to the business plan was there were lots of people with browsers demanding content from servers, and Netscape developed their browser for the sole reason of seeding the market.
I'm not making this up, here are some quotes from the founders:
What killed Netscape is that they overreached themselves; version 1.x of their server products were the first serious products on the market, version 2.x were very good (NES 2.01 is still probably the best webserver I have deployed) but 3.x were terrible: buggy and slow. That's what caused their collapse, people stopped buying their server products. By the time they lost the browser war, it was irrelevant anyway.
Whatever you might feel about McNealy personally, ya gotta give him credit for sheer guts and having razor-sharp focus IMHO
I guess you haven't been following Sun for very long, because McNealy and "focus" don't get along. Sun has flipflopped for years unable to decide if they were a workstation vendor, a server vendor, a vertical one-stop shop a la IBM or something else altogether (see their current marketing campaign around SunONE). The "dot in dotcom" was a disaster for them, and now their sales are being hurt by their own second-hand kit (why buy a brand new Sun Blade workstation when you can pick up a very capable Ultra 60 or 30 for a fraction of the price?). They spent years insisting that Solaris on SPARC was their crown jewels, they fudged the decision on Solaris x86 and now they are selling Linux on x86 machines, this time competing with themselves directly. Or StarOffice - that was part of a plan to compete directly with PCs using cheap workstations running Solaris (the Ultra 5) and thin clients (Sun Ray). It didn't pan out, and they were pretty much forced to give away StarOffice for free (you think they did that out of the goodness of their hearts?). Over the past couple of years their development toolset has been called SPARCworks, SunPro, Forte and now it's SunONE Studio - the same basic tools rebranded depending on whether Sun think the future is embedded, HPC, Internet or "ONE". And Java - for a long time, IBM's JVM on x86 wiped the floor with Sun's on SPARC (may even still be true, haven't checked recently).
Sun have some great products and technologies, but what will kill them is their sheer lack of focus. Right now they need to reestablish the SPARC as the processor for serious computing and establish the Sun Fire/Blade as the platforms for it, and they need to get Java's performance on their low end machines up to the level where it can seriously be used for interactive apps then start selling them in bulk to the people who are currently buying their workstations from Dell.
But the very first thing they need if for McNealy to drop his personal feud with Microsoft and define his company in terms of what his customers want, not simply reacting to whatever MS do.
Same thing with mouse tracking - the X display refresh is not synced in any way with mouse position updates, so moving a window around feels "sticky."
This is actually fairly fundamental to the X paradigm, which is network-oriented. There needs to be no logic on the display device, so things like mouse movements are transmitted over the network for the remote application to see and deal with. It doesn't matter to X if the network is TCP/IP over a WAN or your localhost loopback adaptor. Therefore, it has to be asynchronous, or X would appear to just "freeze" while it was waiting for the network - better that it gives you some feedback and lets you keep working while it sorts itself out than to make you wait for proper synchonization of all GUI updates.
Sun solved this with NeWS which actually does put some interactivity on the display device - for example, you could click a button and the animation of it being "pressed" would be processed on the display and just the button-pressed event sent back over the network (X will have to send mouse-moved and button-pressed events one way, and drawing instructions to display the button pressing over the network). But NeWS never caught on because other workstation vendors weren't prepared to let Sun control the standard.
X is powerful, but it was never made to be fast, and never will be without sacrificing some flexibility. From the Windows perspective the opposite is true: it was made to be fast, and can't be made as flexible as X without sacrificing speed (try using an XP remote desktop over a WAN, for example, or PC Anywhere).
The lady in the booth paused and said, "That's Duane!" She said she recognized my walk.
This is why US special forces can't infiltrate Afghanistan in burkas - they don't walk like women. The British army had a similar problem in Northern Ireland - undercover soldiers tended to march rather than walk, and stood straighter than the laid-back locals, so stood out. So there are definitely applications for this technology, but unfortunately it seems like it would be more useful to terrorists than our side.
also, there were movies made about canada's part in ww2 I've seen them.
Really? What were they aboot, eh?
How amusing. I take it you've never bought anything from a store? How do you suppose it got there, hmmm?
Rail works if you have:
Fortunately, lots of people do have such stuff, or even are such stuff themselves, but over short distances between many points with arbitrary journey start times and routes, mass transit systems are the wrong paradigm.
Sadly, a lot of areas of the US are built in a way that you can't do without a car anymore. I cannot afford to live an area where I can walk to the post office or to a store. But that's not an "advantage".
Sure you can - but you can't expect the benefits of big-city living in a small community. There are plenty of small towns in the US, where people say hello to each other in the streets and the waitress in the diner remembers your name.
But you cannot expect the world to adapt itself to your personal preferences. Either you live in a sprawling metropolis and have access to all it offers, or you live in a small town and adopt the lifestyle. Either way, you make your own choices.
Banning VoIP? Whats next? Possibly banning email to help the USPS?
Yes, what's next, steel tariffs or something? Nah, that'll never happen.
The answer is pretty simple - bad management. Most points that you made (and which I consider completely valid, especially given your impressive experience), can be in fact reduced to bad management. Good management is the exception, not the rule.
I see this asserted time and time again, but experience leads me to the conclusion that if it just as often incompetent workers as incompetent managers that cause projects and hence companies to fail.
For example, management sometimes makes unrealistic claims to customers about what the product can do and when it will be ready, but no more unreasonable or unrealistic than the claims made by programmers to project managers that a particular feature will be ready by a certain date, or even that it will work as it should. Some managers abuse their position and authority by indulging themselves using corporate resources, but no more that prima donna programmers who demand exemptions from the company dress code, free Snapple, aromatherapy etc. Sometimes management demands that the programming staff works long hours, but no more often than the programmers decide to blow off work for the day and play games on the LAN, surf Slashdot, chat on IRC etc.
The true exception is the company where both management and workers have dropped the juvenile "them and us" attitude and work in harmony. There are more than a few of them out there. Usually they are created by the founders not for money per se, but for their own freedom, and are at least their third company. In such a company both management and workers recognise that they have a good thing going here and that it is up to everyone involved to keep it that way.
With MS exchange server, only one copy of the message would be kept and each user would be accessing that message. I wish unix MTAs worked this way.
The Sun for-money one does, I don't think there are any free ones that are that smart, tho'.
I know the lines between the two have begun to blur, now that desktop CPU's have vector instructions (SSE/3DNow); is there something left that desktop PC's (even clusters of them) just don't offer?
You can only horizontally scale a stateless application. For example, you could use a different node to render each scene in a movie - but it would be very difficult to use one machine to render trees throughout the entire movie and another to compute their shadows, because of the interaction between them. Trees create shadows, and shadows fall across trees.
Stateful applications require constant synchronization between computational elements, and for that, you need to vertically scale in something with a nice fast active backplane, otherwise your app will spend more time shuffling data around than working.
I'm looking for a (cheap) solution for filesystem sharing between two linux servers and, since the target is just redundancy, I've come to the following idea
Before you spend a single dollar, ask yourself: if your system is important enough to require fault tolerance, why can't you spend money to get a professional solution? If your system isn't important enough to spend money on, then ordinary bidirectional file replication should be good enough for you. You could do it with rsync and ntpd in a few minutes, for free.
Ever think it might be sort of cool to run that kind of low-fat system on powerful modern hardware? Sort of like an ultimate C64 with, hundreds of megabytes of RAM, and many hundreds of MIPS. Maybe not pretty, but the raw computing horsepower at your fingertips would be awesome, provided you know the system well enough to use it
Couldn't you do this by booting a modern PC into DOS? Running compiled C it would be blazingly fast, your program would get literally 100% of the machine's capacity.
I've seen the acronym "TA" appear in a few articles recently - could someone please explain it for me. Ty
Teaching Assistant. It's common for tenured academic faculty to get their grad students to do the bread-and-butter teaching work, like marking exam papers. TAs are the bottom of the academic pile. It's little better than indentured servitude, but they have to suck it up or get on the bad side of the prof who holds their fate in his hands. TAs are miserable, but everyone's gotta pay their dues.
Of course corporations will work to reduce our salaries, just as we work to raise them.
It depends. Often an employer will pay the highest wages it can, to create a barrier entry to prevent competitors from being able to enter the market.
nyone with half a brain can see management would have us working for free if they had their way
:-)
And who can blame them, when the whole Open Source community is hell-bent on doing so anyway?
If we don't get organized, the few IT jobs left in the US will soon be paying minimum wage
I suggest you read Yourdon's books, "Decline and Fall" and "Rise and Ressurection of the American programmer".
Unions, implemented correctly, start and end with that sentiment. This "rugged individualism" (rugged geekdom?) plays well on TV, but doesn't scale to real life. We've all seen that typical geek skills are becoming more common and less valuable
Answer me this: if your skills are less valuable, by what logic should you even be paid the same for them, let alone more?
There's actually little reason for your skills portfolio to become less valuable over time. Let me give you example. When I started in IT (1996), HTML and Perl CGI were in-demand skills. Very few people had them, lots of people wanted them, those that did have them were well paid for them by those who wanted them. Nowadays, of course, it's difficult to find someone who doesn't have HTML, and Perl CGI isn't really in demand anymore now that PHP, Cold Fusion, JSP et al are used.
But, I'm not unemployed, because I kept my skills up to date (FWIW, I do mission-critical OLTP work for investment banks). Someday, that'll be obsolete too - but I'm not even worried slightly, in 5 years I'll have a whole new skill set, and I'll still be ahead of the curve.
What would a union do? Demand, backed by threats of strikes and pickets, that companies should stick with old-fashioned, labor-intensive technologies to save their members from having to develop their skills? Demand that only union-members are employed, then restrict membership to drive up the price, even though there are people with the same skills happy to do the work for less? That's what happens in unionized industries. It rests on the assumption that all workers are equal and are entitled to equal treatment.
But that falls apart in any business where individuals can have an impact on productivity, and those individuals know that by freely associating, they can get a lot more done. You can see this time and time again - mini-mills outproducing Big Steel is a typical case. The buyers will go to where they can get the most for the least - that's rational economic behavior. Without government protection, customers will leave inefficient union shops in droves. Without customers but the union insisting that the payroll isn't cut, the organization will lose money, eating into its reserves. Eventually those will be gone, and rather than a few people losing their jobs, everyone will lose their jobs. If there are no subsidies forthcoming from the taxpayer, the employer will collapse.
RIAA is the antithesis of free market. RIAA is an organization of media companies who band together to fix prices and shutout competition. RIAA is why you cannot find music published by smaller record labels in music shops.
The RIAA is a cartel, just like De Beers. A cartel is what it's called when sellers in a marketplace agree amongst themselves not to compete with each other on price and often on quality, thereby inflating the price and reducing quality to the lowest common denominator. This is because it is not in the economic interest of any member of the cartel to provide a better quality than any other member, since they won't be rewarded for doing so. A marketplace should act in the best interests of both sellers and buyers, but a cartel skews the market against the buyers. In most regulated markets, cartels are illegal.
Hiring labor is a market like any other. There is a buyer (called an "employer"), and a seller (an "employee"). The seller is selling time, and the buyer is buying it.
The only difference between a cartel and a union is that unions are legal. It is no coincidence that the heavily unionized industries - auto manufacturing, steel, coal mining - have been wrecked, because the internal market that rewards quality and implicity punishes poor quality has been destroyed by the activity of the cartel. A company is an organization for generating economic value. If productivity is not the criteria by which it is measured, instead it employees people to artificially swell the payroll even though they do not produce more than they cost, the cost/benefit calculation fails, and its resources cannot be allocated efficiently. It is doomed therefore to burn its capital, and will implode when that is exhausted. Like any protectionist policy, a union can only protect the interests of its members in the short term, because it is inherently less efficient that a free market in labor.
No - what will happen is that corporations will incorporate trade barriers into the product (such as DVD region coding).
Well, they certainly tried to, but the market responded by creating then satisfying a demand for multi-region players. Without government backing, you can't prevent arbitrage.
given that despite the fact that it's glaringly obvious that UK Brits are systematically ripped off on everything from Cars to Computers compared to our European counterparts - very little action is taken.
Well, duh, who do you suppose is doing the ripping off? Of every pound you spend on petrol, 80p is tax - that's a 400% tax! A pack of cigarettes costs something like 25p to manufacture - and you're paying 15x that in tax!
And everyone pays, even non-drivers - because everything is shipped by road, it just gets factored into the price of loaves of bread. And, despite what the propaganda tells you, the NHS would be well and truly shafted without money from cigarette taxes - the government really wants more smokers, not less. After all they are the perfect citizens: they voluntarily pay more tax their whole lives, then die before they collect their pensions!
Parliament will take no action on rip-off Britain because they are it's biggest fans! And that's why in the past Customs have been so heavy handed with "booze cruisers" - it's only because of recent public outrage and the impossbility of comprehensive enforcement that the rules have recently been (slightly) relaxed.
With that said, I swear to god, multinational cooperations have no conscience. Turn on the news, and all you see is the Enrons, Microsofts, and all these other coopertions who do everything they can to screw the consumer and their employees to make an extra penny. Good for the Europeans, bout damn time someone smacked those companies down, even if it is one with good Karma like nintendo.
On the contrary, multinationals are only operating within the framework provided by national governments. When governments dismantle their trade barriers, such as import tarriffs and quotas, then price differences will simply be arbitraged away by brokers (i.e. you see something selling for $10 in country A and $5 in country B, export/import it and sell it in country A for $6 - eventually the margin will tend to zero). But that can only happen if there are no obstacles to freely moving goods and capital around.
The biggest barrier to this is ironically the EU itself who protect manufacturers like Levi Strauss from UK retailers who source overseas and want to sell at less than Levi's MRRP. Not to mention the distortions the EU create in the market with their subsidies of inefficient industries.
Frankly, I don't know who's worse, corrupt corporations (as distinct from well-run corporations) or corrupt politicians - and the EU isn't even democratically elected! A shareholder has far more influence on a company than a voter has on the European Commission (that's a fact).
Microsoft once (still does?) own a stake of SCO. I remember seeing a copyright Microsoft somewhere either in the OS or in the documentation once.
IIRC, Microsoft sold Xenix to SCO. Xenix was Microsoft Unix running on the 286. Microsoft agreed with AT&T (or someone) not to develop their own Unix.
Microsoft, in all it's *nix hating glory owned a portion of a *nix company
Microsoft don't hate Unix per se; a lot of early MS development (again IIRC) was done on Unix-running DECs cross-compiling to 8086. It's just not their own product, and therefore they have a pathological urge to compete with it.
The big disadvantage to large laser weapons is that they give away their precise position since laser beams travel in perfectly straight lines.
You might think so, but tracer ammo has been around since WW2 and it's still in use today. That suggests that being able to locate a weapon that is firing at you while it is firing isn't as big a tactical advantage as it might first appear.