GPL'ed and is very, very fast. It's been doing very well compared to even the commercial accelerators. In fact the only accelerator that *edges* Turck MMCache in performance is the Zend accelerator. Check at http://php.weblogs.com/php_debugger_cache
for a few numbers and links.
Stick with.NET or J2EE. They're clunky,.NET is expensive, J2EE is slow, but they're both leaps and bounds ahead of PHP.
I can understand why people think that Java is slow, taking that Java GUI widgets are usually slower than native widgets on the desktop. But J2EE is not inheritably slow.
Though there are clunkly J2EE apps out there, I'd wager on a hunch, that the average J2EE app would out-perform a same sized PHP app. If we add a PHP accelerator to the mix, like Turck MMCache, then PHP may have a fighting chance.
But remember, J2EE compiles the web application only *once* at startup, and can also ( and probably does ) optimize for the specific processor that it's running on. PHP, without an accelerator compiles on every hit, and PHP can't optimize for the specific processor unless you have a good sysadmin.
I have wanted to bring my company onto the free/cheap opensource software bandwagon for some time now. And I have the authority to do it. But I always have to consider the issue - can non-techsmart people handle it? Will they be able to open the documents they receive and use them.
Install the open source applications side-by-side with the commercial applications that you plan to replace. You can install Staroffice or OpenOffice right along with MS office, and make sure that it is as support or even more so by your IT department. That means, the users don't get "We don't know, you're on your own" answers from Tech for Staroffice, but then get a dissertation when they ask a question about MS Office.
After you have a installation that is supported ( internally ) and documented as well as ( or better ) than MS Office, ( I know, easier said than done, but do your best ), and the users have had some time to become a little familiar with Staroffice. Start promoting StarOffice as an alternative to MS Office. This could be done easier now. You can direct, for instance, people who need to create simple PDF files to StarOffice for instance. Thats something MS Office can not do without 3rd party software installed.
I think IT departments should give the user an incentive to move to cheaper, well performing software. Eg. The department could get a cut from the money saved, while IT gets a cut and the company holds the rest. This may be difficult to execute because many IT managers don't like decreasing they budget, even if doing so may, in a roundabout way, leave them more money in they pockets at the end of the day.
404 errors are generated by webservers. your browser would return a this page could not be found/resolved page before this was changed.
Right.
And when MS started replacing the web server's 404 error response page with IE's own error page that sent the user to MSN, I didn't hear anyone complain.
Ever noticed that you don't get the default Apache error page when using IE..., ever? IE only displays an error page if it's greater than a certian size. Webmasters have to play tricks like padding the error page with comments so that IE goes ahead and displays it.
To me they are both wrong, but MS has been doing that kind of thing for a while.
What's the point when for $200 the "poor" could by a Linux pc from Wal-Mart.
The simputer is design to be *gasp* simpler then the regular desktop computer.
(i) It was made to to used in areas where electricity is very undependable or non-existant. Where are you going to plug in that walmart PC?
(ii) It was made to be easily shared amongst a large group of people. eg. a co-operative. How are you going to share a desktop amongst a few dozen people? Wouldn't be easier to share a mobile computer?
(iii) It was made to be robust. Are you going to replace walmart's crap keyboad once someone spills something on it?
(iv) It was meant to be financed or bought as an investment. You know, like how westerners buy cars and play for college etc.
You're just going from one pay-for product, to another (albiet less cost). If you REALLY want to show your boss the beauty of alternative software. Show him something thats great, FOR FREE! (that will get any bosses attention).
Of course you can also pay for StarOffice because...
(i) The money going into StarOffice is being used to continue the development OpenOffice, as Sun still pays for a lot of the Development of OpenOffice.
(ii) You can get product support, and training from Sun. Important for even small business, or any overstressed IT department.
Not all of the cost of software is in the purchase of that software.
If Sun were ever to be in a dominant market position, this sort of "bundling" would likely be considered actionably anticompetitive,...
How is this difficult from my school that pays MS $40k/year for site licenses? Staff, faculty, and students get to use a almost all MS software for either free ( faculty and staff ) or a very small fee ( students in some cases).
Sun's stragedy is not much different from many old old site-licenses pricing stragedies. The only thing is that they mandate that the site license be verified with official documentation.
Secondly, visible light is electromagnetic rays. Think about how much of that you absorb in an average day. Augghhhh, the light!! The horrible light! Won't someone think of the children?
Moot point.
It is true light is EM radiation, just as the frequency range cells use, microwave ovens use etc.
But we know that some frequencies are bad for us humans. More specifically the higher freq. above some threshold. We just don't have conclusive evidence what that thresold is.
Many studies, quite a few of them funded by the cell phone industry, insist they do know, and that cell phones are ok.
Personally I agree that more independent work should be done in this area.
Of course SUV drivers don't have a higher survival rate than other cars, they are just more likely to kill others, and not a single bit more likely to survive themselves.
Numbers please. And how were those stats compiled?
And the idea that an Economy car is "undersized" is absolutly silly
As someone who has had to deal with undersized cars his whole life, that comments seems very narrow minded. I'm 6'7, and I have very long legs, there is no way I'm fitting behind the wheel of the average econobox. I've tried.
Frankly in the long run its cheaper and safer for EVERYONE to drive an economy car, and rent a larger vehicle when you NEED one. You know that MAYBE once a year or every two years that you might move, or maybe that weekend or two you actually go camping.
That's the times that *you* need a bigger vehicle. For others it's more convenient and comfortable to have a large car as their primary vehicle.
Run a SQL Server farm on the back end if you cant afford an Oracle license. Don't be an OSS idealogue in the business world, you end up unemployed.
And I would fire the IT guy who causes my company to spend $10,000 for SQL Server in a situation where the free MySQL or Postgres would do.
Just focus on the right tool for the job. If the database is a simple one. If it is regularly backed up and your company can stand a small period of downtime, why on earth would you buy Oracle or MS SQL Server?
This is not to say that MySQL is unreliable. I have *never* seen MySQL crash, or lose any of my data. So it would be silly of me to go with Oracle, just because everyone else is doing it.
You've never heard of "lying by omision" or a "lie of omission"?
In many, many, circumstances omitting important information is a crime.
At any rate, the plaintiff does have a valid argument, if they go in the direction that the RIAA's intent was to mislead. The strength or validity of this argument is up to the courts to decide.
Some people like my dad just want to use the internet, and they don't care how it works, they pay money for an ISP and they expect them to make it work.
I just want to drive my car, I don't care how it works. I don't bother putting in gas, checking my oil or brakes, I just expect it to work.
Kunta
There are basic functionality of any device, that you just have to understand. One has to know that you don't pound the nail with the 'claw' end of the hammer and you don't use a philips screwdriver to undo a screw designed for a flat head.
True the computer needs to be made easier, but you can't just rip essential stuff out in the hopes of making it easier. Your dad may need to access his files from the office one day, and curse the computer because now he has to configure a webserver to do so.
Unix has many scripting languages available to it (Perl, Python, TCL, in addition to bash, tch, et. al.), and at least one of them will usually be installed by default on a modern system. They can script command line apps like nobody's business, but unfortunately the more modern GUI apps provide much less scriptablity. CORBA and Kparts might help this if they were more commonly available. This is an area Unix-like environments (including MacOS) need to really improve in.
Right.
That's a separate issue though. Unix needs a universally accepted binary component protocol. We need one badly.
Traditionally we've always used libraries to share functionality amongst seperate programs. But libraries make it difficult to make decisions at runtime, and discovery of shared libraries at runtime is primative at best ( LD_LIBRARY_PATH, etc. ). Most component archs/protocols offer a registry for finding exported interfaces, ie. functionality, at runtime. The application just needs to know the version of the interface it requires. eg. the application asks 'I need SpellChecker Interface v 0.5, does anyone provide that?' A KDE module may respond, a Java module may respond or a GNOME module may respond.
I believe GNOME with bonobo, KDE with kparts, and JCP with Java's Corba/RMI need to get together and work out a baseline common binary component standard. That's wishful thinking, granted, but doing so would do the linux community a world of good.
How many times have Linus or others said that the goal for Linux is NOT to attack Microsoft's monopoly...
It would be hard to argue that Linus et al. are not gunning for Microsoft, BSD, and other competing operating systems.
In fact, if we be honest about it, I think it is obvious that they are. Not that there's anything work with being competitive.
Remember when someone published a report that IIS on WinNT was insanely faster than Apache on Linux? Linus and the other kernel hackers had kHTTPD, and a ton on fixes in the kernel in no time flat. A bunch more fixes/enhancements made it into the next major kernel release as well.
I think the large number of complaints is that although source-compatibility is -basically- maintained, you still have to recompile your apps. One of the nice things Microsoft has done is that you don't have to recompile your Win32-based app to work in.NET -- well, not completely, anyhow. This does have the side effect of dirtying up the API a bit. So it's a trade-off. Backwards compatibility does make GUIs easier for people to adopt -- who wants to constantly have to download new apps to work with the latest version???
What on earth are you talking about? I think you should clarify your post a little, cause sounds like you're saying.NET is source compatible with Win32 API or MFC.
.NET is in no way, shape or form source compatible with Win23 API or MFC. Anyone who's spent five minutes looking at the.NET Foundation Class Libraries would know this.
.NET runtime will have absolutely nothing to do with a Win32 app or MFC based application. Not unless you're just pointing out that MS is currently supporting both Win32 and.NET platforms, so you can run those apps side-by-side.
So yeah, you can pull in unmanaged routines using.NET's 'Marshal','Interop' functions, but do not expect to pull in your entire MFC or Win32 gui int a.NET application that way.
And plus your entire GUI event pump is different, so I don't see that working well for a gui anyway. Using the interop functions is like using JNI in Java to pull in C/C++ code, I'd say. Not the most elegant way to do things.
Moving from MFC, or worse yet plain Win32 to.NET is a major rewrite. Many of us have to do it cause you know MS is going to make it very difficult to stay. Already documentation in the new MSDNs is leaning heavy to.NET. But hey, it's their prerogative right? I mean my years of investment learning MFC/Win32 API doesn't mean anything right? Not that I'm not bitter or anything.
I worked on moving a non-gui MFC based parser class to managed C++, and that was a lot of work to keep things on the CLR at all times. It was like we basically ported to a new platform.
Here we have Scott McNealy telling people ""Don't touch open-source software unless you have a team of intellectual-property lawyers prepared to scour every single piece" of open-source code. " yet they're also releasing an open sourced distribution of Linux.
A lesson I learnt early on in my business dealings is that you sometimes have to work closely with people you do not trust or see entirely eye-to-eye with. And that you sometimes have to bury the hachet long enough for the two of you to get some common good done.
McNealy, Sun, open-source organizations, supporters have a lot in common; For starts a competitor that has 90+% of markets they're interested in.
Linux only has at most 3% of the desktop market. Now let me ask you, do you have a better way to get a significant piece of the other 97% of the desktop market? A make it a plan that you're in a position to directly execute or initiate, by the way.
If you don't, I think we should just tell Sun thanks, and move on.
The.Net framework and development tools are also free, and can be downloaded. And efforts are being made for making.Net portable. Examples include Microsoft's own Rotor (Windows, FreeBSD and Mac OSX), Mono (Linux) and DotGNU (Linux).
Language is nothing. The dependabilty and completeness of the libraries is really what's important.
These projects may one day produces a sturdy C# compiler or CIL interpreter. But how complete will their libraries be?
The Microsoft FCL library is huge. Before we can say that linux supports.NET we're going to have to have a complete and dependable FCL implementation.
That FCL implemenation will also have to behave very closely to the Microsoft's library for compatibility. I don't think Mono, or DotGNU is going to be able to pull that off and I suspect MS realizes this as well.
No. How much security does NFS have built-in? Exactly none
Care to back that up?
NFS protocol has built in encryption/authentication using GSS-API since version 3. That was quite a few years ago. NFS version 4 is out.
I maintained a lab running on an encrypted NFS FS about 3 years ago, on Solaris 7.
Linux didn't have support for encrypted NFS because the kernel hackers couldn't get encryption into the kernel at the time. Now that 2.6 has kernel encryption services Linux will support the full NFSv4 spec. Or at least support the security features.
But you can't blame the engineers that developed NFS, they've had encryption/authentication built into the protocol for years now.
That kit allows you to run any code that you want to anyway. Plus getting one allows companies to see that there is a paying group of individuals that would like configurable/extensible electronic products.
It's funny that many people criticize the software and media industry for promoting DRM and DMCA type laws, but then the same people turn around and promote/utilize cracks like this.
What do you expect the companies to do? Sit there and watch this happen?
There are already MPlayer, the ffmpeg library, mjpegtools, bbmpeg, Ogg Vorbis and Theora, Cinelerra... I for one don't feel that I need a bone thrown to me by Real, much less a proprietary, binary-only, NDA-encumbered (no, more like encrusted) one.
What are you going to use for streaming? How do you encode? I hope you plan to pay royalities on that MPEG video. You are talking about MPEG4 right?... right?
Helix is giving as a complete end-to-end multimedia solution. Client, encoder, and server. All under a single project ran by a company with at least a bit of clout in the multimedia software industry.
Helix is much needed by the open source community. Wake-up. MPlayer's legality is questionable at best. Video encoding/decoding is a patent minefield. With the number of codecs MPlayer does natively, I'm surprised that they haven't got in trouble as yet. Plus they use MS dlls for a bunch of others.
How do we push *that* as the 'de-factor' Linux player?... and have linux taken seriously?
Ogg is nice but they have very little clout in the industry. Look how long they've been trying to take a foothold with vorbis? Now how long till the unfinished theora, becomes widespread??
Slashdot Mirror
Try Turck MMCache instead.
GPL'ed and is very, very fast. It's been doing very well compared to even the commercial accelerators. In fact the only accelerator that *edges* Turck MMCache in performance is the Zend accelerator. Check at http://php.weblogs.com/php_debugger_cache for a few numbers and links.
I can understand why people think that Java is slow, taking that Java GUI widgets are usually slower than native widgets on the desktop. But J2EE is not inheritably slow.
Though there are clunkly J2EE apps out there, I'd wager on a hunch, that the average J2EE app would out-perform a same sized PHP app. If we add a PHP accelerator to the mix, like Turck MMCache, then PHP may have a fighting chance.
But remember, J2EE compiles the web application only *once* at startup, and can also ( and probably does ) optimize for the specific processor that it's running on. PHP, without an accelerator compiles on every hit, and PHP can't optimize for the specific processor unless you have a good sysadmin.
Install the open source applications side-by-side with the commercial applications that you plan to replace. You can install Staroffice or OpenOffice right along with MS office, and make sure that it is as support or even more so by your IT department. That means, the users don't get "We don't know, you're on your own" answers from Tech for Staroffice, but then get a dissertation when they ask a question about MS Office.
After you have a installation that is supported ( internally ) and documented as well as ( or better ) than MS Office, ( I know, easier said than done, but do your best ), and the users have had some time to become a little familiar with Staroffice. Start promoting StarOffice as an alternative to MS Office. This could be done easier now. You can direct, for instance, people who need to create simple PDF files to StarOffice for instance. Thats something MS Office can not do without 3rd party software installed.
I think IT departments should give the user an incentive to move to cheaper, well performing software. Eg. The department could get a cut from the money saved, while IT gets a cut and the company holds the rest. This may be difficult to execute because many IT managers don't like decreasing they budget, even if doing so may, in a roundabout way, leave them more money in they pockets at the end of the day.
Right.
And when MS started replacing the web server's 404 error response page with IE's own error page that sent the user to MSN, I didn't hear anyone complain.
Ever noticed that you don't get the default Apache error page when using IE..., ever? IE only displays an error page if it's greater than a certian size. Webmasters have to play tricks like padding the error page with comments so that IE goes ahead and displays it.
To me they are both wrong, but MS has been doing that kind of thing for a while.
The simputer is design to be *gasp* simpler then the regular desktop computer.
(i) It was made to to used in areas where electricity is very undependable or non-existant. Where are you going to plug in that walmart PC?
(ii) It was made to be easily shared amongst a large group of people. eg. a co-operative. How are you going to share a desktop amongst a few dozen people? Wouldn't be easier to share a mobile computer?
(iii) It was made to be robust. Are you going to replace walmart's crap keyboad once someone spills something on it?
(iv) It was meant to be financed or bought as an investment. You know, like how westerners buy cars and play for college etc.
thanks for pointing that out, you're right. you're right.
I believe microwave radiation used for communication utilizes a much higher frequency than visible light.
Of course you can also pay for StarOffice because...
(i) The money going into StarOffice is being used to continue the development OpenOffice, as Sun still pays for a lot of the Development of OpenOffice.
(ii) You can get product support, and training from Sun. Important for even small business, or any overstressed IT department.
Not all of the cost of software is in the purchase of that software.
How is this difficult from my school that pays MS $40k/year for site licenses? Staff, faculty, and students get to use a almost all MS software for either free ( faculty and staff ) or a very small fee ( students in some cases).
Sun's stragedy is not much different from many old old site-licenses pricing stragedies. The only thing is that they mandate that the site license be verified with official documentation.
Moot point.
It is true light is EM radiation, just as the frequency range cells use, microwave ovens use etc.
But we know that some frequencies are bad for us humans. More specifically the higher freq. above some threshold. We just don't have conclusive evidence what that thresold is.
Many studies, quite a few of them funded by the cell phone industry, insist they do know, and that cell phones are ok.
Personally I agree that more independent work should be done in this area.
Numbers please. And how were those stats compiled?
And the idea that an Economy car is "undersized" is absolutly silly
As someone who has had to deal with undersized cars his whole life, that comments seems very narrow minded. I'm 6'7, and I have very long legs, there is no way I'm fitting behind the wheel of the average econobox. I've tried.
Frankly in the long run its cheaper and safer for EVERYONE to drive an economy car, and rent a larger vehicle when you NEED one. You know that MAYBE once a year or every two years that you might move, or maybe that weekend or two you actually go camping.
That's the times that *you* need a bigger vehicle. For others it's more convenient and comfortable to have a large car as their primary vehicle.
Supposedly should be out by now.
And I would fire the IT guy who causes my company to spend $10,000 for SQL Server in a situation where the free MySQL or Postgres would do.
Just focus on the right tool for the job. If the database is a simple one. If it is regularly backed up and your company can stand a small period of downtime, why on earth would you buy Oracle or MS SQL Server?
This is not to say that MySQL is unreliable. I have *never* seen MySQL crash, or lose any of my data. So it would be silly of me to go with Oracle, just because everyone else is doing it.
The right tool for the job people.
You've never heard of "lying by omision" or a "lie of omission"?
In many, many, circumstances omitting important information is a crime.
At any rate, the plaintiff does have a valid argument, if they go in the direction that the RIAA's intent was to mislead. The strength or validity of this argument is up to the courts to decide.
There are basic functionality of any device, that you just have to understand. One has to know that you don't pound the nail with the 'claw' end of the hammer and you don't use a philips screwdriver to undo a screw designed for a flat head.
True the computer needs to be made easier, but you can't just rip essential stuff out in the hopes of making it easier. Your dad may need to access his files from the office one day, and curse the computer because now he has to configure a webserver to do so.
Right.
That's a separate issue though. Unix needs a universally accepted binary component protocol. We need one badly.
Traditionally we've always used libraries to share functionality amongst seperate programs. But libraries make it difficult to make decisions at runtime, and discovery of shared libraries at runtime is primative at best ( LD_LIBRARY_PATH, etc. ). Most component archs/protocols offer a registry for finding exported interfaces, ie. functionality, at runtime. The application just needs to know the version of the interface it requires. eg. the application asks 'I need SpellChecker Interface v 0.5, does anyone provide that?' A KDE module may respond, a Java module may respond or a GNOME module may respond.
I believe GNOME with bonobo, KDE with kparts, and JCP with Java's Corba/RMI need to get together and work out a baseline common binary component standard. That's wishful thinking, granted, but doing so would do the linux community a world of good.
You can use the site or download and install a single university version.
I've used it in the past, and was about to reinstall and promote our site. It works well.
It would be hard to argue that Linus et al. are not gunning for Microsoft, BSD, and other competing operating systems.
In fact, if we be honest about it, I think it is obvious that they are. Not that there's anything work with being competitive.
Remember when someone published a report that IIS on WinNT was insanely faster than Apache on Linux? Linus and the other kernel hackers had kHTTPD, and a ton on fixes in the kernel in no time flat. A bunch more fixes/enhancements made it into the next major kernel release as well.
This is just one example.
What on earth are you talking about? I think you should clarify your post a little, cause sounds like you're saying .NET is source compatible with Win32 API or MFC.
.NET is in no way, shape or form source compatible with Win23 API or MFC. Anyone who's spent five minutes looking at the .NET Foundation Class Libraries would know this.
.NET runtime will have absolutely nothing to do with a Win32 app or MFC based application. Not unless you're just pointing out that MS is currently supporting both Win32 and .NET platforms, so you can run those apps side-by-side.
So yeah, you can pull in unmanaged routines using .NET's 'Marshal','Interop' functions, but do not expect to pull in your entire MFC or Win32 gui int a .NET application that way.
And plus your entire GUI event pump is different, so I don't see that working well for a gui anyway. Using the interop functions is like using JNI in Java to pull in C/C++ code, I'd say. Not the most elegant way to do things.
Moving from MFC, or worse yet plain Win32 to .NET is a major rewrite. Many of us have to do it cause you know MS is going to make it very difficult to stay. Already documentation in the new MSDNs is leaning heavy to .NET. But hey, it's their prerogative right? I mean my years of investment learning MFC/Win32 API doesn't mean anything right? Not that I'm not bitter or anything.
I worked on moving a non-gui MFC based parser class to managed C++, and that was a lot of work to keep things on the CLR at all times. It was like we basically ported to a new platform.
...Which really is what it is.
A lesson I learnt early on in my business dealings is that you sometimes have to work closely with people you do not trust or see entirely eye-to-eye with. And that you sometimes have to bury the hachet long enough for the two of you to get some common good done.
McNealy, Sun, open-source organizations, supporters have a lot in common; For starts a competitor that has 90+% of markets they're interested in.
Linux only has at most 3% of the desktop market. Now let me ask you, do you have a better way to get a significant piece of the other 97% of the desktop market? A make it a plan that you're in a position to directly execute or initiate, by the way.
If you don't, I think we should just tell Sun thanks, and move on.
Language is nothing. The dependabilty and completeness of the libraries is really what's important.
These projects may one day produces a sturdy C# compiler or CIL interpreter. But how complete will their libraries be?
The Microsoft FCL library is huge. Before we can say that linux supports .NET we're going to have to have a complete and dependable FCL implementation.
That FCL implemenation will also have to behave very closely to the Microsoft's library for compatibility. I don't think Mono, or DotGNU is going to be able to pull that off and I suspect MS realizes this as well.
My 2cents.
Care to back that up?
NFS protocol has built in encryption/authentication using GSS-API since version 3. That was quite a few years ago. NFS version 4 is out.
I maintained a lab running on an encrypted NFS FS about 3 years ago, on Solaris 7.
Linux didn't have support for encrypted NFS because the kernel hackers couldn't get encryption into the kernel at the time. Now that 2.6 has kernel encryption services Linux will support the full NFSv4 spec. Or at least support the security features.
But you can't blame the engineers that developed NFS, they've had encryption/authentication built into the protocol for years now.
You can already run Linux on the playstation by paying for the PS2 Linux kit at http://playstation2-linux.com/
That kit allows you to run any code that you want to anyway. Plus getting one allows companies to see that there is a paying group of individuals that would like configurable/extensible electronic products.
It's funny that many people criticize the software and media industry for promoting DRM and DMCA type laws, but then the same people turn around and promote/utilize cracks like this.
What do you expect the companies to do? Sit there and watch this happen?
Yea. Try explaining to the brass that it's going to cost you $350,000 and it's free software.
The software is still free. If you choose to hire Red Hat to support your network, you pay them.
It's that simple.
What are you going to use for streaming? How do you encode? I hope you plan to pay royalities on that MPEG video. You are talking about MPEG4 right?... right?
Helix is giving as a complete end-to-end multimedia solution. Client, encoder, and server. All under a single project ran by a company with at least a bit of clout in the multimedia software industry.
Helix is much needed by the open source community. Wake-up. MPlayer's legality is questionable at best. Video encoding/decoding is a patent minefield. With the number of codecs MPlayer does natively, I'm surprised that they haven't got in trouble as yet. Plus they use MS dlls for a bunch of others.
How do we push *that* as the 'de-factor' Linux player?... and have linux taken seriously?
Ogg is nice but they have very little clout in the industry. Look how long they've been trying to take a foothold with vorbis? Now how long till the unfinished theora, becomes widespread??