Re:A few things that could lead to scalability
on
On PHP and Scaling
·
· Score: 3, Insightful
Sorry, your abbreviations are confusing me. DFS? I know Disk File System and Distributed File System, but neither of those seem a good fit for what you're talking about. So... what are you talking about?
Actually, the bundling started with IE 3 IIRC, but towards the end of its lifecycle
The bundling started, I think, with IE 2 which I think got bundled with NT4. It might also have been on some early copies of 95. It was upgraded to IE 3 pretty quickly, though (in OSR1?). A few people are still using IE3, because it's what they got when they bought Win95.
I don't like using java on the client. It's mostly slow and unresponsive.
That's mainly the fault of AWT's poor event handling model. Java programs that are written with interfaces built ground-up without using AWT tend to be just as responsive as C/C++ programs, and much more responsive than applications whose UI is built in a scripting language.
I do the opposite. I write in a mixture of Java and native code that only runs on Windows. It could theoretically be ported to Linux, but I haven't done it yet.
I do this because (1) I like working in an object-oriented, garbage-collected, runtime-typed language, and Java was the most convenient for me when I started all this (I'd probably use C# & either.NET or mono if I was starting now, but I have a load of reusable code I don't want to have to port), and (2) I need to use native APIs to achieve the effets I want to. Java's OS interface just wasn't up to the job.
Doing this has given me a pretty unique insight into Java's shortcomings. It's this: AWT. The system is an abomination.
And before you say "Swing!", anything that is built as a layer on top of AWT is doomed from the start, in my opinion. AWT is badly designed and inefficiently implemented, and nothing can save it IMO.
I've heard it said that it's becoming very difficult to acquire Java (or at least Old Brown Java, the most well known variety) these days. Apparently, the Japanese are buying so much of it, that the rest of have to make do with Aged Timor instead. It makes me wonder if what I buy from my supplier as OBJ really is. Still, there isn't much difference in quality between the two, so I can't really complain, even if I could find out they were wrong.
$ time java -jar g:/usr/local/lib/jakarta-tomcat-3.2.3/lib/ant.jar Searching for build.xml... Could not locate a build file!
real 0m1.268s
Not brilliant, I guess. But that's starting it over a network connection on a 400MHz processor, using cygwin's bash which ain't exactly fast to start with. For comparison's sake:
$ time make make: *** No targets specified and no makefile found. Stop.
real 0m0.353s
not to mention has about the worst syntax for a user-edited configuration file possible.
Java is not the best for command line programs mainly because VM initialization is expensive (in terms of time). This could be possibly alleviated by having a system-level VM that was initialized at boot time.
Or using an ahead of time compiler, which is included by default in most modern Linux distributions.
And of the remaining 12 items, 2 are purely administrative projects (foundation and conferences), and one (incubator) is a collection of early-stage projects, an equally high proportion of which are using Java.
For an organisation that started with a flagship project that was written in C, Apache certainly does a lot of work in Java.
Because the optimal data rate over the line uses all available bandwidth in it. The lossy compression is optimised to encode the most important information into that bandwidth, which is something a simple analogue channel cannot do. Therefore, while the same amount of information is transmitted in both cases (according to the theory), the information transmitted with well-designed lossy compression is more important information and therefore produces a better quality signal.
I'm sorry I can't put this in formal terms, my mathematics isn't quite up to it.
Err, no. No matter how good the line is, sending a well-designed lossily-compressed digital signal over it at the maximum bitrate that can be supported by the line is guaranteed to give you a better result than sending an analogue signal over it. Information theory requires it.
The plan's based on the idea that if you have multiple telescopes you can use interferometry to make them behave similarly to a single telescope that has a diameter the size of the gap between them.
I think (a) you'd need more than 2 of them for this to work, and (b) beyond a few light minutes' distance the communications lag would make it unworkable, but an interesting idea nonetheless.
The problem is that our sample is currently heavily biased toward gas giants because of detection techniques that have been used so far.
Also, life-supporting planets is only one factor of many in the Drake equation. Others are _much_ more contentious, like proportion on which life arises, and proportion of life bearing planets that give rise to civilization. Disagreements on those two tend can be in large orders of magnitude.
Unless this observation was done in a wavelength atenuated by atmosphere (e.g. in near UV)
Actually, if they're going to try to determine atmospheric composition, I think they might _have_ to use a wavelength attenuated by atmosphere (?? I'm not very up on spectroscopic techniques, but that sounds logical).
Ah, thank you. You put me in mind to go and look up this:
Bernard Woolley: "Now, may I just have your approval for this Local Government Allowances Amendment Number 2 for this year's regulations." Jim Hacker: "What is it?" Bernard Woolley: "It is a Statutory Instrument to be laid before the House. As Minister responsible for local government we need you to authorize that the revised Paragraph 5 of Number 2 Regulations 1971 shall come into operation on March the 18th next, revoking Regulation 7 of the Local Government Allowances Amendment Regulations 1954(b)." Jim Hacker: "What the hell does all that mean?" Bernard Woolley: "It is all right, there is an explanatory note, Minister. These Regulations are to make provisions for prescribing the amount of attendance and financial loss allowances payable to the members of local authorities. Explanatory note: Regulation 3 of the Local Government Allowances Amendment Regulation 1971 ("the 1971 regulations") substituted a new regulation for Regulation 3 of the 1954 Regulations. Regulation 3 of the Local Government Allowances Amendment Regulation 1972 ("the 1972 regulations") further made amends Regulation 3 of the 1954 Regulations by increasing the maximum rates of attendance and financial loss allowances. Regulation 7 of the 1972 Regulations revoked both regulation 3 and 5 of the 1971 Regulations, Regulation 5 being a regulation revoking earlier spent regulations with the effect from 1st April next. These regulations preserve Regulations 2 and 5 of the 1971 Regulations by revoking Regulation 7 of the 1972 Regulations. Jim Hacker: "And that's an explanatory note???"
You have to remember that the EU can only exist right now because there's a consensus between the states that its better for it to exist than for it not to exist. This means that the system must be designed to allow for this kind of thing, otherwise member states might feel trapped by it, and decide they're better off without it.
Somebody needs to write an OS where the windowing operations are all done in the memory allocator. Wouldn't that be the more efficient way to go about it?
I considered it, but decided the virtual memory management subsystem was the more logical place.
Note: this is actually serious -- in the design for my OS kernel, which may or may not ever cease being utter vapourware, the VMM implements support for memory mapped windows whose data are transparently copied on modification (with possible transformation, e.g. scaling, rotation, clipping, alpha blending) into a larger window. This can be used to implement a memory-mapped GUI system reasonably efficiently. I was going to have it occurring in user space, but decided that for efficiency reasons it would _have_ to happen in the kernel (a test implementation of the idea using Linux framebuffers and standard userspace VM features was too slow to be usable).
I think everyone complaining about missing the point is missing another one: Windows isn't _supposed_ to do these things out of the box. It can't do half of the things that almost everyone uses it for (e.g. word processing) without additional tools.
Windows is a base system thay you add what you want to. It always has been (although it is moving away from this position these days, with more and more integrated features).
Permission thunking between NTFS ACLs and Unix-style perms slows it all down quite a bit
Actually, I don't think that's what makes cygwin slow -- I think the problem is that Windows doesn't support fork() so it has to be emulated in user-mode software, which is a bit of a PITA.
You're comparing a system whose (original) philosophy is "provide basic tools needed by everyone and let people add additional capabilities with 3rd party applications" with one which is "provide everything anyone is ever likely to need on one set of CDs and give them a huge menu asking what to install at the start".
Of course the former (Windows) isn't going to be as capable as the latter (most Linux distributions) if you don't use it the way it's supposed to be used.
You might as well say you can't do word processing on Windows because WordPad's so shite.
It seems wildid issues only S/MIME e-mail certificates. This company also offers SSL (https) secure server certificates, which is much more useful. They also seem to be significantly better at what they do. I wouldn't trust a certificate issuer who has an expired certificate on their own secure server. What's their excuse for having an expired self-signed cert???
It prevents man-in-the-middle attacks. That's the most important reason for me to use a trusted CA.
A self-signed cert can prevent man-in-the-middle attacks if you have a copy of the cert downloaded.
Most CAs these days only verify that an e-mail they send to the domain in question is received by the applicant, the same as this lot. I got one from Equifax on this basis, and they're trusted by default in both IE and Mozilla.
Slashdot Mirror
Sorry, your abbreviations are confusing me. DFS? I know Disk File System and Distributed File System, but neither of those seem a good fit for what you're talking about. So... what are you talking about?
Actually, the bundling started with IE 3 IIRC, but towards the end of its lifecycle
The bundling started, I think, with IE 2 which I think got bundled with NT4. It might also have been on some early copies of 95. It was upgraded to IE 3 pretty quickly, though (in OSR1?). A few people are still using IE3, because it's what they got when they bought Win95.
Wow, I don't expect him to last long there making public statements like this!
Didn't you know? "We used to have problems but its all OK now" is a perfectly acceptable company promotional line. Even top executives use it.
I don't like using java on the client. It's mostly slow and unresponsive.
That's mainly the fault of AWT's poor event handling model. Java programs that are written with interfaces built ground-up without using AWT tend to be just as responsive as C/C++ programs, and much more responsive than applications whose UI is built in a scripting language.
I do the opposite. I write in a mixture of Java and native code that only runs on Windows. It could theoretically be ported to Linux, but I haven't done it yet.
.NET or mono if I was starting now, but I have a load of reusable code I don't want to have to port), and (2) I need to use native APIs to achieve the effets I want to. Java's OS interface just wasn't up to the job.
I do this because (1) I like working in an object-oriented, garbage-collected, runtime-typed language, and Java was the most convenient for me when I started all this (I'd probably use C# & either
Doing this has given me a pretty unique insight into Java's shortcomings. It's this: AWT. The system is an abomination.
And before you say "Swing!", anything that is built as a layer on top of AWT is doomed from the start, in my opinion. AWT is badly designed and inefficiently implemented, and nothing can save it IMO.
I've heard it said that it's becoming very difficult to acquire Java (or at least Old Brown Java, the most well known variety) these days. Apparently, the Japanese are buying so much of it, that the rest of have to make do with Aged Timor instead. It makes me wonder if what I buy from my supplier as OBJ really is. Still, there isn't much difference in quality between the two, so I can't really complain, even if I could find out they were wrong.
I've used 'ant' which is very slow to start up
...
:)
$ time java -jar g:/usr/local/lib/jakarta-tomcat-3.2.3/lib/ant.jar
Searching for build.xml
Could not locate a build file!
real 0m1.268s
Not brilliant, I guess. But that's starting it over a network connection on a 400MHz processor, using cygwin's bash which ain't exactly fast to start with. For comparison's sake:
$ time make
make: *** No targets specified and no makefile found. Stop.
real 0m0.353s
not to mention has about the worst syntax for a user-edited configuration file possible.
I ain't going to argue there!
Java is not the best for command line programs mainly because VM initialization is expensive (in terms of time). This could be possibly alleviated by having a system-level VM that was initialized at boot time.
Or using an ahead of time compiler, which is included by default in most modern Linux distributions.
And of the remaining 12 items, 2 are purely administrative projects (foundation and conferences), and one (incubator) is a collection of early-stage projects, an equally high proportion of which are using Java.
For an organisation that started with a flagship project that was written in C, Apache certainly does a lot of work in Java.
Because the optimal data rate over the line uses all available bandwidth in it. The lossy compression is optimised to encode the most important information into that bandwidth, which is something a simple analogue channel cannot do. Therefore, while the same amount of information is transmitted in both cases (according to the theory), the information transmitted with well-designed lossy compression is more important information and therefore produces a better quality signal.
I'm sorry I can't put this in formal terms, my mathematics isn't quite up to it.
Err, no. No matter how good the line is, sending a well-designed lossily-compressed digital signal over it at the maximum bitrate that can be supported by the line is guaranteed to give you a better result than sending an analogue signal over it. Information theory requires it.
They're aliens. They'd probably prefer Windows Media Audio v1. Somebody has to.
The plan's based on the idea that if you have multiple telescopes you can use interferometry to make them behave similarly to a single telescope that has a diameter the size of the gap between them.
I think (a) you'd need more than 2 of them for this to work, and (b) beyond a few light minutes' distance the communications lag would make it unworkable, but an interesting idea nonetheless.
Kafka! Prepare for metamorphosis!
The problem is that our sample is currently heavily biased toward gas giants because of detection techniques that have been used so far.
Also, life-supporting planets is only one factor of many in the Drake equation. Others are _much_ more contentious, like proportion on which life arises, and proportion of life bearing planets that give rise to civilization. Disagreements on those two tend can be in large orders of magnitude.
Unless this observation was done in a wavelength atenuated by atmosphere (e.g. in near UV)
Actually, if they're going to try to determine atmospheric composition, I think they might _have_ to use a wavelength attenuated by atmosphere (?? I'm not very up on spectroscopic techniques, but that sounds logical).
Hubble will, presumably, be left up there.
:)
Does this mean salvage rights are available?
Ah, thank you. You put me in mind to go and look up this:
Bernard Woolley: "Now, may I just have your approval for this Local Government Allowances Amendment Number 2 for this year's regulations."
Jim Hacker: "What is it?"
Bernard Woolley: "It is a Statutory Instrument to be laid before the House. As Minister responsible for local government we need you to authorize that the revised Paragraph 5 of Number 2 Regulations 1971 shall come into operation on March the 18th next, revoking Regulation 7 of the Local Government Allowances Amendment Regulations 1954(b)."
Jim Hacker: "What the hell does all that mean?"
Bernard Woolley: "It is all right, there is an explanatory note, Minister. These Regulations are to make provisions for prescribing the amount of attendance and financial loss allowances payable to the members of local authorities. Explanatory note: Regulation 3 of the Local Government Allowances Amendment Regulation 1971 ("the 1971 regulations") substituted a new regulation for Regulation 3 of the 1954 Regulations. Regulation 3 of the Local Government Allowances Amendment Regulation 1972 ("the 1972 regulations") further made amends Regulation 3 of the 1954 Regulations by increasing the maximum rates of attendance and financial loss allowances. Regulation 7 of the 1972 Regulations revoked both regulation 3 and 5 of the 1971 Regulations, Regulation 5 being a regulation revoking earlier spent regulations with the effect from 1st April next. These regulations preserve Regulations 2 and 5 of the 1971 Regulations by revoking Regulation 7 of the 1972 Regulations.
Jim Hacker: "And that's an explanatory note???"
From Yes, Minister (of course).
You have to remember that the EU can only exist right now because there's a consensus between the states that its better for it to exist than for it not to exist. This means that the system must be designed to allow for this kind of thing, otherwise member states might feel trapped by it, and decide they're better off without it.
Somebody needs to write an OS where the windowing operations are all done in the memory allocator. Wouldn't that be the more efficient way to go about it?
I considered it, but decided the virtual memory management subsystem was the more logical place.
Note: this is actually serious -- in the design for my OS kernel, which may or may not ever cease being utter vapourware, the VMM implements support for memory mapped windows whose data are transparently copied on modification (with possible transformation, e.g. scaling, rotation, clipping, alpha blending) into a larger window. This can be used to implement a memory-mapped GUI system reasonably efficiently. I was going to have it occurring in user space, but decided that for efficiency reasons it would _have_ to happen in the kernel (a test implementation of the idea using Linux framebuffers and standard userspace VM features was too slow to be usable).
I think everyone complaining about missing the point is missing another one: Windows isn't _supposed_ to do these things out of the box. It can't do half of the things that almost everyone uses it for (e.g. word processing) without additional tools.
Windows is a base system thay you add what you want to. It always has been (although it is moving away from this position these days, with more and more integrated features).
Permission thunking between NTFS ACLs and Unix-style perms slows it all down quite a bit
Actually, I don't think that's what makes cygwin slow -- I think the problem is that Windows doesn't support fork() so it has to be emulated in user-mode software, which is a bit of a PITA.
Apples are crunchy.
Oranges are sweet.
Hmmmm.
You're comparing a system whose (original) philosophy is "provide basic tools needed by everyone and let people add additional capabilities with 3rd party applications" with one which is "provide everything anyone is ever likely to need on one set of CDs and give them a huge menu asking what to install at the start".
Of course the former (Windows) isn't going to be as capable as the latter (most Linux distributions) if you don't use it the way it's supposed to be used.
You might as well say you can't do word processing on Windows because WordPad's so shite.
Learn to live in the real world, will you?
It seems wildid issues only S/MIME e-mail certificates. This company also offers SSL (https) secure server certificates, which is much more useful. They also seem to be significantly better at what they do. I wouldn't trust a certificate issuer who has an expired certificate on their own secure server. What's their excuse for having an expired self-signed cert???
It prevents man-in-the-middle attacks. That's the most important reason for me to use a trusted CA.
A self-signed cert can prevent man-in-the-middle attacks if you have a copy of the cert downloaded.
Most CAs these days only verify that an e-mail they send to the domain in question is received by the applicant, the same as this lot. I got one from Equifax on this basis, and they're trusted by default in both IE and Mozilla.