The problem with "frameworks" that appear now -- Java, all generations of COM,.NET, etc. is that they basically don't do anything useful. They are implementations of someone's "vision" how software is supposed to be developed, and how existing two component models -- that is, one provided by shared libraries, and another one by interprocess communications, should be replaced by something more "object oriented". And all those visions are wrong, simply because at this point mankind has as much chances to develop a better component model as thousand years ago it had chances to develop better means of long-range transportation than a horse and a sailing ship -- the level of technology was inadequate for that and will be for quite a while.
But people are ambitious and instead of writing little useful things -- say, a serializer for data formats that will simplify communications, they are trying to create a GREAT NEW SYSTEM with everything existing replaced with some GREAT NEW IDEA plus every petty piece of idiosyncaasy that a particular developer or company has at the moment. That produced Java, and that plus typical Microsoft's aspirations to create more incompatibilities per hour than they were able before, produced.NET.
The fact is, for a programmer who is writing a new code, "framework" like this gives very little -- dynamic libraries existed for a long time, anything written in C can be now called from any other language that can load a library, and IPC is still not used at its full potential, especially considering that programmers (Gnome ones included) were indoctrinated with superstition that context switches are more expensive than additional in-core pages while in modern systems context switches are cheap, IPC is efficient, but data-structures bloat that causes huge amount of data sitting in memory in a tangled mess becomes a huge problem for speed even if memory itself is cheap.
For a programmer who wishes to use components developed for Windows it gives nothing, too -- the whole strategy of Windows development is to keep things unusable outside Windows, and certainly mixing few proprietary pieces in the mix will poison everything developed on Windows just like things being developed now are poisoned for everyone who agrees to anything less than a complete emulation of every ugly piece that Microsoft managed to throw into Windows. In other words -- if you will want to run something developed for Windows, you will need Windows, all Windows and you will have very little chance to add anything but Windows. You can emulate it, but then you will basically become it. So all this talk about "inevitable success" of.NET on Windows is pointless -- at best you can outrun them (despite some umm... strange ideology changes that reminds me Communist Party line, the development of GNOME is going fast enough), at least you can ignore them (Microsoft will replace it with something else soon anyway), but you can't assimilate them.
Oh, btw., the use of borg-style metaphors was intentional but not essential to this piece of advice.
He is saying something that 1. isn't exactly what he is doing, 2. what can be interpreted by journalists and PHBs (certainly was interpreted by Slashdot editors and hordes of local trolls) in a way that he is not expecting Linux to be used on a desktop at all. If he wants to clearly explain that he sees desktop as a "territory" that a system has to hold to keep Microsoft from taking over everything else, he MUST SAY SO EXACTLY IN THOSE WORDS. If he want to say that he does not want to play "monopoly takes all" game on a desktop he should better just keep this to himself because the mentality of the public who learned to look at Microsoft with dollar-sign shaped eyes is not ready to accept such an idea yet.
Contrary to popular belief, Windows is not "ready" for everyday person's desktop either -- in fact the usability of Windows-base desktop sucks at the extent of being ridiculous. It's just there because Microsoft placed it on every desktop computer.
I have used Linux on my desktops exclusively for eight years, and have not encountered any problem that is worse than ones that "everyday person" encounters while using Windows.
C# has no useful for games networking capabilities whatsoever -- it's based on the idea of a remote (procedure,method,...) call while networked games have to deal with asyncronous streams of requests and messages. One can reduce those streams (plus all related structures that describe which stream describes which instances of objects, and how their state should propagate between clients and servers) to calls but that would be like writing a GUI on a Turing machine.
And how conveniently it was to omit the fact that IE was starting to come preinstalled with the OS right in the middle of this process, and it was exactly that time when users who upgraded their systems or bought a new one, got a browser with it and stopped downloading Netscape.
I've seen formatting problems on both word processors. I've exported work and seen problems rendering it under Word.
INSTALL THE FONTS THAT WERE USED WHEN DOCUMENTS WERE COMPOSED!!!
Grrr... How many times do I have to tell people that? Microsoft formats (as opposed to, say, TeX) are completely enslaved to font sizes, so any kind of fallback font causes formatting problems simply because it's not the same size. What does everybody expect, that TrueType Arial will magically appear on every X server without a sysadmin bothering to install it?
What about PowerPoint?
Same as with Word
What about Outlook?
Masochists that need Outlook-like interface and bloat reproduced to the finest details, are welcome to use Evolution, everyone else can use pine or mutt -- I do.
What about Access, FrontPage, or any of the other apps available under the Office family of products?
Those apps should not exist in the first place.
You haven't actually used anything but Office for longer than it takes to read a Microsoft ad in a glossy magazine, did you?
If ya cant just dubbel klik it and and make it go with flashy bits and stuff, then you just lost 95% of your users.
I don't care about 95% of the users -- Windows, if it was sold in boxes and not preinstalled, would be unavailable for 95% of users, too. The point is, in this case stupid user is at fault, and he should shut up.
I can use pine (and only pine!) for my mailbox _precisely_ because it can't sort messages -- mutt can, and this is why I can't use it. My mailbox is currently >200000 messages long, and nothing, absolutely nothing, can read it remotely except a combination of cyrus on the server and pine at the client. Pine happily reads it at usable speed with no significant delays over everything from 100Mbps Ethernet to 14.4Kbps cellular phone, and selections/zooming/coloring functions allow me to handle my email easily and efficiently.
Because he uses broken charset and has a hotmail address, therefore he is obviously a Windows-only users, and his opinion on Linux games is absolutely irrelevant.
What kind of performance hit do you take when you run a graphics intensive game under X?
Same as under any other platform if you use a hardware-accelerated OpenGL.
I don't play games on my Linux box but I have noticed certain applications that crawl under Linux but perform very well in Windows or MacOS, even with lesser hardware. Flash and Blender just to name a few.
Your card is running as 2d-only because it's either unsupported, or you are too lazy to install an OpenGL driver -- and probably is a piece of junk because everything non-junk has some decent driver already. As for Flash, its Linux port simply sucks because Macromedia knows about Linux as much as I know about the inhabitants' of the Andromeda Nebula taste in paintings.
If there were a large difference in the value of a dollar "within" the US versus "outside" of the US then there would be a massive amount of currency moving from the place where it was valued to where it was not. People would move the dollar, buy some goods and then import or export the goods. This is in fact what happens in countries where the official external value for a dollar is different than the internal value.
And this does not happen precisely because the monopoly for "globalization" owned by large corporations (who do exactly what you have described) is protected by isolationism for everyone else.
What does that mean? Let's step away from dollars for a second. If you take an ounce of gold it can buy much more (in terms of real estate, labour or most material products) in a poor country than in a rich company.
Probably because gold currently has almost no value by itself -- in most of countries its very limited local use in electronics and jewelry can't justify its price, therefore the price of gold itself is tied to the common currency, that happens to be a dollar.
If one will use some more usable way to measure the price (say, land located in similar conditions) it would become apparent that dollar as internal currency of US and dollar as an international currency have completely different value. This difference in value can be decreased if true globalization taken place -- when consumers and procucers globally could trade with reduced barriers. However what is called "globalization" now is globalization for large corporations plus isolationism for everyone else -- a large company can choose in what country it wants to produce something, a small company usually has much less choice, and most of individuals have absolutely no choice, in what country to work.
Of course, if people will more freely move between countries, dollar will cease being split to dollar-in-US and dollar-outside-US, and Americans will lose this advantage. However that also will mean that other countries will finally get a chance to improve their economy without having to use overpriced US currency for international transactions, what now simply sucks products into US under inadequate prices, thus suffocating the development of true global economy.
Some RPC-2 drives can be converted into RPC-1 ones by changing their firmware -- for example, BDV212B, that I have seen being sold for $59. And free software players -- say, Ogle that I use now, don't care about those things.
If so, how much is the rest of your life worth? Maybe I would prefer to pay it to you now rather than to have you around all that time.
The truth is, time isn't worth money, time is the most expensive and irreplaceable thing a human may have. Money may worth or not worth the time spent on them, but this is a different story.
And, no, Gnumeric is nowhere near up to it. On my machine, Gnumeric takes longer to load a 3-sheet workbook than XL does to load a 50-sheet workbook where each sheet is at least twice as large. It also feels slow; XL tends to feel punchy and responsive.
Don't run it on 486, with pixmap theme, IDE hard drive with UDMA disabled, and Trident 8900 VGA. Because this is the only way to get the results you have described.
Userbase has nothing to do with it -- StarOffice is not a financial application and must not be used for this purpose, along with MS Office and its likes. Custom-written banking software, with likely userbase of one, however, is.
What Gnumeric? the whole idea of using a spreadsheet application for anything banking-related, other than for displaying tables prepared by other programs, is a problem. Spreadsheets are unreliable and not protected against inconsistency in their scripts, and a choice of the most common spreadsheet only confirms that people who use them have no clue.
...that not only ??AA executives believe that everyone should license everything from them only in some "copy-protected" form, the only way to actually make those restrictions actually restrict anything is by licensing some piece of shit from Microsoft?
2. Open source programmers simply don't care if AOL will make more or less money off their work. Their problem is completely different -- they have to have a standards-compliant browser on non-Microsoft platforms that they happen to use, and the only viable project is still Mozilla -- Opera and Konqueror have their inherent limitations precisely because they became faster by ignoring or breaking cumbersone standards, and there is no reasonable way to fix the standard in the situation where W3C has basically idiots+Microsoft at the helm. Mozilla, by following W3C madness such as DOM and countless stylesheets models, allows people such as myself to keep Microsoft products away from their desktops, and that goal has nothing to do with AOL or any other company's money.
FastCGI is better than just a bunch of symmetric processes, however it has some serious flaws -- among them poor security model for processes that run on other hosts (fhttpd reverses the logins, backends' connect to the server, and those connections authenticate on the server), and a need to proxy the response through a server for processes that run locally (fhttpd passes a client's fd to the backend process).
PHP already does connections pooling -- it doesn't help at all, IIS still has to be multithreaded. On Windows multithreading is necessary because of poor system design (out of process COM servers suck), so Microsoft engineers had no better choice anyway, but on Unix multiple processes allow to make more clean and secure design without any noticeable overhead, so there is no need to use multithreading unless it's a developer's goal to use exclusively multithreaded monsters like Java.
Slashdot Mirror
The problem with "frameworks" that appear now -- Java, all generations of COM, .NET, etc. is that they basically don't do anything useful. They are implementations of someone's "vision" how software is supposed to be developed, and how existing two component models -- that is, one provided by shared libraries, and another one by interprocess communications, should be replaced by something more "object oriented". And all those visions are wrong, simply because at this point mankind has as much chances to develop a better component model as thousand years ago it had chances to develop better means of long-range transportation than a horse and a sailing ship -- the level of technology was inadequate for that and will be for quite a while.
But people are ambitious and instead of writing little useful things -- say, a serializer for data formats that will simplify communications, they are trying to create a GREAT NEW SYSTEM with everything existing replaced with some GREAT NEW IDEA plus every petty piece of idiosyncaasy that a particular developer or company has at the moment. That produced Java, and that plus typical Microsoft's aspirations to create more incompatibilities per hour than they were able before, produced .NET.
The fact is, for a programmer who is writing a new code, "framework" like this gives very little -- dynamic libraries existed for a long time, anything written in C can be now called from any other language that can load a library, and IPC is still not used at its full potential, especially considering that programmers (Gnome ones included) were indoctrinated with superstition that context switches are more expensive than additional in-core pages while in modern systems context switches are cheap, IPC is efficient, but data-structures bloat that causes huge amount of data sitting in memory in a tangled mess becomes a huge problem for speed even if memory itself is cheap.
For a programmer who wishes to use components developed for Windows it gives nothing, too -- the whole strategy of Windows development is to keep things unusable outside Windows, and certainly mixing few proprietary pieces in the mix will poison everything developed on Windows just like things being developed now are poisoned for everyone who agrees to anything less than a complete emulation of every ugly piece that Microsoft managed to throw into Windows. In other words -- if you will want to run something developed for Windows, you will need Windows, all Windows and you will have very little chance to add anything but Windows. You can emulate it, but then you will basically become it. So all this talk about "inevitable success" of .NET on Windows is pointless -- at best you can outrun them (despite some umm... strange ideology changes that reminds me Communist Party line, the development of GNOME is going fast enough), at least you can ignore them (Microsoft will replace it with something else soon anyway), but you can't assimilate them.
Oh, btw., the use of borg-style metaphors was intentional but not essential to this piece of advice.
He is saying something that 1. isn't exactly what he is doing, 2. what can be interpreted by journalists and PHBs (certainly was interpreted by Slashdot editors and hordes of local trolls) in a way that he is not expecting Linux to be used on a desktop at all. If he wants to clearly explain that he sees desktop as a "territory" that a system has to hold to keep Microsoft from taking over everything else, he MUST SAY SO EXACTLY IN THOSE WORDS. If he want to say that he does not want to play "monopoly takes all" game on a desktop he should better just keep this to himself because the mentality of the public who learned to look at Microsoft with dollar-sign shaped eyes is not ready to accept such an idea yet.
Contrary to popular belief, Windows is not "ready" for everyday person's desktop either -- in fact the usability of Windows-base desktop sucks at the extent of being ridiculous. It's just there because Microsoft placed it on every desktop computer.
I have used Linux on my desktops exclusively for eight years, and have not encountered any problem that is worse than ones that "everyday person" encounters while using Windows.
...run an HTTP server on it. Don't tell anyone that it's there. What will you see?
In my case it was few hundreds of Code Red requests from few tens of hosts per day.
C# has no useful for games networking capabilities whatsoever -- it's based on the idea of a remote (procedure,method, ...) call while networked games have to deal with asyncronous streams of requests and messages. One can reduce those streams (plus all related structures that describe which stream describes which instances of objects, and how their state should propagate between clients and servers) to calls but that would be like writing a GUI on a Turing machine.
There are only three causes of insecure code:
1. Developers' ignorance.
2. Developers' stupidity.
3. Selling underdeveloped software.
And how conveniently it was to omit the fact that IE was starting to come preinstalled with the OS right in the middle of this process, and it was exactly that time when users who upgraded their systems or bought a new one, got a browser with it and stopped downloading Netscape.
I've seen formatting problems on both word processors. I've exported work and seen problems rendering it under Word.
INSTALL THE FONTS THAT WERE USED WHEN DOCUMENTS WERE COMPOSED!!!
Grrr... How many times do I have to tell people that? Microsoft formats (as opposed to, say, TeX) are completely enslaved to font sizes, so any kind of fallback font causes formatting problems simply because it's not the same size. What does everybody expect, that TrueType Arial will magically appear on every X server without a sysadmin bothering to install it?
What about PowerPoint?
Same as with Word
What about Outlook?
Masochists that need Outlook-like interface and bloat reproduced to the finest details, are welcome to use Evolution, everyone else can use pine or mutt -- I do.
What about Access, FrontPage, or any of the other apps available under the Office family of products?
Those apps should not exist in the first place.
You haven't actually used anything but Office for longer than it takes to read a Microsoft ad in a glossy magazine, did you?
If ya cant just dubbel klik it and and make it go with flashy bits and stuff, then you just lost 95% of your users.
I don't care about 95% of the users -- Windows, if it was sold in boxes and not preinstalled, would be unavailable for 95% of users, too. The point is, in this case stupid user is at fault, and he should shut up.
I can use pine (and only pine!) for my mailbox _precisely_ because it can't sort messages -- mutt can, and this is why I can't use it. My mailbox is currently >200000 messages long, and nothing, absolutely nothing, can read it remotely except a combination of cyrus on the server and pine at the client. Pine happily reads it at usable speed with no significant delays over everything from 100Mbps Ethernet to 14.4Kbps cellular phone, and selections/zooming/coloring functions allow me to handle my email easily and efficiently.
Because he uses broken charset and has a hotmail address, therefore he is obviously a Windows-only users, and his opinion on Linux games is absolutely irrelevant.
What kind of performance hit do you take when you run a graphics intensive game under X?
Same as under any other platform if you use a hardware-accelerated OpenGL.
I don't play games on my Linux box but I have noticed certain applications that crawl under Linux but perform very well in Windows or MacOS, even with lesser hardware. Flash and Blender just to name a few.
Your card is running as 2d-only because it's either unsupported, or you are too lazy to install an OpenGL driver -- and probably is a piece of junk because everything non-junk has some decent driver already. As for Flash, its Linux port simply sucks because Macromedia knows about Linux as much as I know about the inhabitants' of the Andromeda Nebula taste in paintings.
Not everybody can set oversea production, or move to another country to work.
If there were a large difference in the value of a dollar "within" the US versus "outside" of the US then there would be a massive amount of currency moving from the place where it was valued to where it was not. People would move the dollar, buy some goods and then import or export the goods. This is in fact what happens in countries where the official external value for a dollar is different than the internal value.
And this does not happen precisely because the monopoly for "globalization" owned by large corporations (who do exactly what you have described) is protected by isolationism for everyone else.
What does that mean? Let's step away from dollars for a second. If you take an ounce of gold it can buy much more (in terms of real estate, labour or most material products) in a poor country than in a rich company.
Probably because gold currently has almost no value by itself -- in most of countries its very limited local use in electronics and jewelry can't justify its price, therefore the price of gold itself is tied to the common currency, that happens to be a dollar.
If one will use some more usable way to measure the price (say, land located in similar conditions) it would become apparent that dollar as internal currency of US and dollar as an international currency have completely different value. This difference in value can be decreased if true globalization taken place -- when consumers and procucers globally could trade with reduced barriers. However what is called "globalization" now is globalization for large corporations plus isolationism for everyone else -- a large company can choose in what country it wants to produce something, a small company usually has much less choice, and most of individuals have absolutely no choice, in what country to work.
Of course, if people will more freely move between countries, dollar will cease being split to dollar-in-US and dollar-outside-US, and Americans will lose this advantage. However that also will mean that other countries will finally get a chance to improve their economy without having to use overpriced US currency for international transactions, what now simply sucks products into US under inadequate prices, thus suffocating the development of true global economy.
Some RPC-2 drives can be converted into RPC-1 ones by changing their firmware -- for example, BDV212B, that I have seen being sold for $59. And free software players -- say, Ogle that I use now, don't care about those things.
My time is worth money;
If so, how much is the rest of your life worth? Maybe I would prefer to pay it to you now rather than to have you around all that time.
The truth is, time isn't worth money, time is the most expensive and irreplaceable thing a human may have. Money may worth or not worth the time spent on them, but this is a different story.
And, no, Gnumeric is nowhere near up to it. On my machine, Gnumeric takes longer to load a 3-sheet workbook than XL does to load a 50-sheet workbook where each sheet is at least twice as large. It also feels slow; XL tends to feel punchy and responsive.
Don't run it on 486, with pixmap theme, IDE hard drive with UDMA disabled, and Trident 8900 VGA. Because this is the only way to get the results you have described.
Userbase has nothing to do with it -- StarOffice is not a financial application and must not be used for this purpose, along with MS Office and its likes. Custom-written banking software, with likely userbase of one, however, is.
What Gnumeric? the whole idea of using a spreadsheet application for anything banking-related, other than for displaying tables prepared by other programs, is a problem. Spreadsheets are unreliable and not protected against inconsistency in their scripts, and a choice of the most common spreadsheet only confirms that people who use them have no clue.
I work at a bank. We have mean excel workbooks, with 20+ worksheets, dozens of vba scripts, links to other workbooks
So, what is the name of that bank? I don't want to ever trust my money to people who are going to use Excel with vba scripts to handle them.
...that not only ??AA executives believe that everyone should license everything from them only in some "copy-protected" form, the only way to actually make those restrictions actually restrict anything is by licensing some piece of shit from Microsoft?
1. You are a moron.
2. Open source programmers simply don't care if AOL will make more or less money off their work. Their problem is completely different -- they have to have a standards-compliant browser on non-Microsoft platforms that they happen to use, and the only viable project is still Mozilla -- Opera and Konqueror have their inherent limitations precisely because they became faster by ignoring or breaking cumbersone standards, and there is no reasonable way to fix the standard in the situation where W3C has basically idiots+Microsoft at the helm. Mozilla, by following W3C madness such as DOM and countless stylesheets models, allows people such as myself to keep Microsoft products away from their desktops, and that goal has nothing to do with AOL or any other company's money.
FastCGI is better than just a bunch of symmetric processes, however it has some serious flaws -- among them poor security model for processes that run on other hosts (fhttpd reverses the logins, backends' connect to the server, and those connections authenticate on the server), and a need to proxy the response through a server for processes that run locally (fhttpd passes a client's fd to the backend process).
Other than that, FastCGI is a good idea.
PHP already does connections pooling -- it doesn't help at all, IIS still has to be multithreaded. On Windows multithreading is necessary because of poor system design (out of process COM servers suck), so Microsoft engineers had no better choice anyway, but on Unix multiple processes allow to make more clean and secure design without any noticeable overhead, so there is no need to use multithreading unless it's a developer's goal to use exclusively multithreaded monsters like Java.