By the way - how much of Google or Yahoo is written in Java... let's see - none of it.
Every site makes their own choice of tools and technologies. Just because Google and Yahoo doesn't use Java, doesn't mean we can conclude it is useless
Also, where does the "Gig of RAM per web page view" figure come from?
No, you have to profile first. Profiling will tell you whether there is even any point in tuning, and, if so, what goals are reasonable.
It's a classic chicken and egg conundrum... If your program meets your performance requirements, why spend time profiling in the first place (but perhaps this is always necessary with Java apps).
I still believe that premature optimization is way too prevalent, unnecessary, and problematic. I recommend the following approach:
Make the program function correctly first
If it doesnt meet performance requirements, profile it to find out where the problems are
If necessary, optimize code highlighted by profiler
Yeah... ATI now makes the "Flipper" chip that powers the Gamecube graphics engine. They obtained this technology when they bought the company (Artx) that was making it. This is not to be mistaken for the current Radeon graphics card technology. It is way different.
Perforce has a link on their website for a paper entitled Web Content Management with Perforcefrom a conference in 1999. This is slightly later than the patent filing date, but no doubt the product info was public before then since they have been in existence since 1996.
They can say whatever they want in the description sections, but it is the claims sections that are the important parts of the patent. Claim 1 is very specifically confined to Web site content. All other claims are based on this one.
I bet the only way they could get this issued, was to make it this specific in the claims
From the patent: 1. A system for file management for files containing website content comprising:...
The very first claim (upon which all other claims are based), indicates that it is strictly for files containing website content.
Even if this patent doesnt get shot down in future, it is not possible for them to extend their claims to affect other products (CVS,ClearCase,Perforce,MKS,PVCS,RCS,etc..) for more general revision control purposes
The filing data was Feb. 4, 1998. The first claim looks so weak as to be achievable by a large number of products by this date. Alas, I guess it must be challenged legally now that it has been issued.
I belive PharLap was one of the first pioneers of dos extenders (companies like QEMM used them). Their 386-Dos extender was first released in 1986. I believe one of the original extended memory protocols was called VCPI. The DPMI protocol was done later on, with a number of companies, such as Rational, and Microsoft participating in its development and standardization.
Interestingly enough, Microsoft supported DPMI in Win95 (Dos box/Dos mode), but Microsoft never ever implemented the full DPMI v1.0 spec. They only ever implemented DPMI 0.9. Some believed, that in doing so they helped encourage the transition of applications to be rewritten natively for Windows
Dont be concerned with pagefile size as much as with page faults. Run the performance monitor, select Process (and choose your game process), then select page faults/sec to look at the rate of page faults as you play.
"Page Faults/sec is the rate Page Faults occur in the threads executing in this process. A page fault occurs when a thread refers to a virtual memory page that is not in its working set in main memory. This will not cause the page to be fetched from disk if it is on the standby list and hence already in main memory, or if it is in use by another process with whom the page is shared."... these page faults are costly reads from disk
What is your point?... You have just described a disk intensive operation, not just a ram intensive one... of course your hard drives will be working hard...
I bet if OpenGL 2.0 were ready, he would be able to render his peroxide natively.
Re:Formalised features of Perl (in this book?)
on
XML and Perl
·
· Score: 1
As you're also aware, most Comp Sci courses fawn over Pascal, a VERY formalized language. However, it's not mentioned much past education circles (and Apple afficionados).
Ever heard of Borland's Delphi product? The language is object-Pascal.
If you're into fiction then I would suggest you try some of Umberto Eco's novels
While I would agree that his writing is exceptionally good (even considering it is translated from Italian to English). I would hardly categorize the first 2 books you mentioned as Science Fiction or Fantasy. They are usually characterized as "historical fiction".
As an aside, Foucaults Pendulum was the most difficult book I have ever gotten through. It was packed with so much detail on religious history, I almost had an aneurism trying to get through it all. I have just started reading hiis new one "Baudolino which I am enjoying so far
I havent seen Steven Erikson mentioned, so here goes.
He has 4 books out that are interconnected, but not necessarily a series. Fantastic read! I have read the first 3 (Gardens of the moon, Deadhouse gates, Memories of Ice) and am itching to read the 4th which was just published. These are a bit more on the fantasy side with a bit of (soft) sf.
On the harder sf side, Vernor Vinge is excellent.
I also enjoyed Neil Stephenson... it is hard to compare his various books because the genres were so different, but I enjoyed Cryptonomicon more than Snow Crash and Diamond Age, because I think the writing is better. Your mileage may vary
Software engineering is a practical craft. No amount of process will ever make up for proper training, proper documentation, proper version control, and proper testing. Ever. And that's the way it is. If you have good people, set them free. If you don't, spend a little money to train them to their highest potential instead of trying to make them good cogs in a crappy buzzword wheel.
This is quite a simplistic view of the development world. I think it "may" have some relevance for smaller projects, but a recipe for disaster in anything approaching a large project size (i.e. hundreds of people involved).
Just look at some ongoing sizeable Open Source projects (where you might expect no process) as examples. Lets take Mozilla: There are guidelines for coding, guidelines for checking source in, guidelines for reporting/fixing bugs, guidelines for deciding what gets into a particular release, etc... This is a process. It may not be the the process-du-jour, but it exists. It has to at this scope, otherwise anarchy results.
Has anything really changed in the field of science since the "Betrayers of the Truth" book was written?
Ok, maybe the people holding the purse strings, but that is about it. The pressure has always been there to "publish or perish" in the academic scientific community.
Now, why is that? Well, it was all about grants. When I worked in academia (pre 1989), the grants were doled out by the various funding bodies. Their criteria was based on what you had published (seemingly) at least as much as what valid science you proposed. It was very much an old boys network.
An example is in order here: The group I was with, was putting in for a 5 year multi-million dollar operating grant. A good portion of the proposal had some very weak science, but was included largely by the political power of the department head (against the sound judgement of many lower technical peons). The funding body sent a site visit team, who were made up of renowned researchers in the various areas of expertise. The review went okay except for the "weak science" area, which got lambasted by the reviewers... the net result was that the grant was not funded. Sounds correct so far (the system was working), right?...
Well, "our" group leaders got together to post-mortem the review and decide how to proceed. So what did they do? Examine the weak science? NO! They proceeded to critique a bunch of the examiners by looking at their CV's and how many papers they had published. Then they formulated an appeal on this basis, trying to influence as many of their high level contacts as possible. Net result: They got a grant (thankfully, they left the weak science bit out).
I left academia shortly thereafter...
Now I ask, has anything changed? I am not so sure?
The 1.2 release also had a build problem where some of the changes checked into the branch did not get pulled for the build.
Specifically, from an artical on www.mozillazine.org they say that "the 1.2 release tag was not complete so builds created from that tag may have additional problems"
I suspect that the difference may be due to this
Re:WARNING - online banking likely to fail -WRONG
on
Mozilla 1.2 Unleashed
·
· Score: 2, Informative
WRONG!
I have been using 1.2 versions for ages without any problems with online banking.
Check your preferences for enabling cookies.
If you look at the bug again, you will notice it is now marked WORKSFORME (and indicating that it was a user settings issue).
I went to a presentation by Ballard Power about 10 years ago regarding their fuel cell technology. I recall at the time one of the speakers saying that they needed someone to invent a a hydrogen "sponge" to really move things forward into the mainstream quickly. Since this has yet to happen (amongst other issues), things are proceeding at a much slower pace.
I am not trying to dispute that Ada is a good language for critical safety related software... but it is only as good as the people and methodology/process being employed.
Consider the fact that the code for the Ariane 5 rocket which crashed because of a software problem, was written in Ada.
"The widest precision thatâ(TM)s not too slow on odayâ(TM)s most nearly ubiquitous âoeWintelâ computers is not double (8 bytes wide, 53 sig. bits) but IEEE 754 double extended or long double (Â10 bytes wide, 64 sig. bits). This is the format in which all local scalar variables should be declared, in which all anonymous variables should be evaluated by default. C99 would permit this (not require it, alas), but â¦
Microsoftâ(TM)s compilers for Windows NT, 2000, ⦠disable that format.
Java disallows it.
Most ANSI C, C++ and Fortran compilers spurn it.
( Appleâ(TM)s SANE got it right for 680x0-based Macs, but lost it upon switching to Power-Macs.)"
Slashdot Mirror
By the way - how much of Google or Yahoo is written in Java... let's see - none of it.
Every site makes their own choice of tools and technologies. Just because Google and Yahoo doesn't use Java, doesn't mean we can conclude it is useless
Also, where does the "Gig of RAM per web page view" figure come from?
No, you have to profile first. Profiling will tell you whether there is even any point in tuning, and, if so, what goals are reasonable.
It's a classic chicken and egg conundrum... If your program meets your performance requirements, why spend time profiling in the first place (but perhaps this is always necessary with Java apps).
I still believe that premature optimization is way too prevalent, unnecessary, and problematic. I recommend the following approach:
Make the program function correctly first
If it doesnt meet performance requirements, profile it to find out where the problems are
If necessary, optimize code highlighted by profiler
- support for yEnc
This is covered by the following Bugzilla RFE bug entry:
http://bugzilla.mozilla.org/show_bug.cgi?id=1199 64
(you will have to cut and paste it in your browser as Bugzilla rejects direct links from Slashdot)
Someone has posted an initial patch to add this feature, but it is not in the 1.4a release
Yeah... ATI now makes the "Flipper" chip that powers the Gamecube graphics engine. They obtained this technology when they bought the company (Artx) that was making it. This is not to be mistaken for the current Radeon graphics card technology. It is way different.
Perforce has a link on their website for a paper entitled Web Content Management with Perforcefrom a conference in 1999. This is slightly later than the patent filing date, but no doubt the product info was public before then since they have been in existence since 1996.
They can say whatever they want in the description sections, but it is the claims sections that are the important parts of the patent. Claim 1 is very specifically confined to Web site content. All other claims are based on this one.
I bet the only way they could get this issued, was to make it this specific in the claims
From the patent: 1. A system for file management for files containing website content comprising: ...
The very first claim (upon which all other claims are based), indicates that it is strictly for files containing website content.
Even if this patent doesnt get shot down in future, it is not possible for them to extend their claims to affect other products (CVS,ClearCase,Perforce,MKS,PVCS,RCS,etc..) for more general revision control purposes
The filing data was Feb. 4, 1998. The first claim looks so weak as to be achievable by a large number of products by this date. Alas, I guess it must be challenged legally now that it has been issued.
I belive PharLap was one of the first pioneers of dos extenders (companies like QEMM used them). Their 386-Dos extender was first released in 1986. I believe one of the original extended memory protocols was called VCPI. The DPMI protocol was done later on, with a number of companies, such as Rational, and Microsoft participating in its development and standardization.
Interestingly enough, Microsoft supported DPMI in Win95 (Dos box/Dos mode), but Microsoft never ever implemented the full DPMI v1.0 spec. They only ever implemented DPMI 0.9. Some believed, that in doing so they helped encourage the transition of applications to be rewritten natively for Windows
Dont be concerned with pagefile size as much as with page faults. Run the performance monitor, select Process (and choose your game process), then select page faults/sec to look at the rate of page faults as you play.
"Page Faults/sec is the rate Page Faults occur in the threads executing in this process. A page fault occurs when a thread refers to a virtual memory page that is not in its working set in main memory. This will not cause the page to be fetched from disk if it is on the standby list and hence already in main memory, or if it is in use by another process with whom the page is shared."... these page faults are costly reads from disk
What is your point?...
You have just described a disk intensive operation, not just a ram intensive one... of course your hard drives will be working hard...
I bet if OpenGL 2.0 were ready, he would be able to render his peroxide natively.
As you're also aware, most Comp Sci courses fawn over Pascal, a VERY formalized language. However, it's not mentioned much past education circles (and Apple afficionados).
Ever heard of Borland's Delphi product? The language is object-Pascal.
If you're into fiction then I would suggest you try some of Umberto Eco's novels
While I would agree that his writing is exceptionally good (even considering it is translated from Italian to English). I would hardly categorize the first 2 books you mentioned as Science Fiction or Fantasy. They are usually characterized as "historical fiction".
As an aside, Foucaults Pendulum was the most difficult book I have ever gotten through. It was packed with so much detail on religious history, I almost had an aneurism trying to get through it all. I have just started reading hiis new one "Baudolino which I am enjoying so far
I havent seen Steven Erikson mentioned, so here goes.
He has 4 books out that are interconnected, but not necessarily a series. Fantastic read! I have read the first 3 (Gardens of the moon, Deadhouse gates, Memories of Ice) and am itching to read the 4th which was just published. These are a bit more on the fantasy side with a bit of (soft) sf.
On the harder sf side, Vernor Vinge is excellent.
I also enjoyed Neil Stephenson... it is hard to compare his various books because the genres were so different, but I enjoyed Cryptonomicon more than Snow Crash and Diamond Age, because I think the writing is better. Your mileage may vary
Software engineering is a practical craft. No amount of process will ever make up for proper training, proper documentation, proper version control, and proper testing. Ever. And that's the way it is. If you have good people, set them free. If you don't, spend a little money to train them to their highest potential instead of trying to make them good cogs in a crappy buzzword wheel.
This is quite a simplistic view of the development world. I think it "may" have some relevance for smaller projects, but a recipe for disaster in anything approaching a large project size (i.e. hundreds of people involved).
Just look at some ongoing sizeable Open Source projects (where you might expect no process) as examples. Lets take Mozilla: There are guidelines for coding, guidelines for checking source in, guidelines for reporting/fixing bugs, guidelines for deciding what gets into a particular release, etc... This is a process. It may not be the the process-du-jour, but it exists. It has to at this scope, otherwise anarchy results.
Has anything really changed in the field of science since the "Betrayers of the Truth" book was written?
Ok, maybe the people holding the purse strings, but that is about it. The pressure has always been there to "publish or perish" in the academic scientific community.
Now, why is that? Well, it was all about grants. When I worked in academia (pre 1989), the grants were doled out by the various funding bodies. Their criteria was based on what you had published (seemingly) at least as much as what valid science you proposed. It was very much an old boys network.
An example is in order here: The group I was with, was putting in for a 5 year multi-million dollar operating grant. A good portion of the proposal had some very weak science, but was included largely by the political power of the department head (against the sound judgement of many lower technical peons). The funding body sent a site visit team, who were made up of renowned researchers in the various areas of expertise. The review went okay except for the "weak science" area, which got lambasted by the reviewers... the net result was that the grant was not funded. Sounds correct so far (the system was working), right?... Well, "our" group leaders got together to post-mortem the review and decide how to proceed. So what did they do? Examine the weak science? NO! They proceeded to critique a bunch of the examiners by looking at their CV's and how many papers they had published. Then they formulated an appeal on this basis, trying to influence as many of their high level contacts as possible. Net result: They got a grant (thankfully, they left the weak science bit out).
I left academia shortly thereafter...
Now I ask, has anything changed? I am not so sure?
The 1.2 release also had a build problem where some of the changes checked into the branch did not get pulled for the build.
Specifically, from an artical on www.mozillazine.org they say that "the 1.2 release tag was not complete so builds created from that tag may have additional problems"
I suspect that the difference may be due to this
WRONG!
I have been using 1.2 versions for ages without any problems with online banking.
Check your preferences for enabling cookies.
If you look at the bug again, you will notice it is now marked WORKSFORME (and indicating that it was a user settings issue).
http://bugzilla.mozilla.org/show_bug.cgi?id=1720 97
my 5.x server has 724 days and counting.
Moderate this down... Its a spam link to a distasteful image (your mileage may vary).
I went to a presentation by Ballard Power about 10 years ago regarding their fuel cell technology. I recall at the time one of the speakers saying that they needed someone to invent a a hydrogen "sponge" to really move things forward into the mainstream quickly. Since this has yet to happen (amongst other issues), things are proceeding at a much slower pace.
I guess it wasnt obvious enough that it was intended as a joke... sigh.
I was under the impression that they hid the CLI so that they could justify selling (Visual) Basic to the masses.
I am not trying to dispute that Ada is a good language for critical safety related software... but it is only as good as the people and methodology/process being employed.
Consider the fact that the code for the Ariane 5 rocket which crashed because of a software problem, was written in Ada.
It's not the hardware that is holding back IEEE754 properly, but rather the compilers
According to W. Kahan (one of the fathers of IEEE754) (see this link to PDF article)
"The widest precision thatâ(TM)s not too slow on odayâ(TM)s most nearly ubiquitous âoeWintelâ computers is not double (8 bytes wide, 53 sig. bits) but IEEE 754 double extended or long double (Â10 bytes wide, 64 sig. bits). This is the format in which all local scalar variables should be declared, in which all anonymous variables should be evaluated by default. C99 would permit this (not require it, alas), but â¦Microsoftâ(TM)s compilers for Windows NT, 2000, ⦠disable that format.
Java disallows it.
Most ANSI C, C++ and Fortran compilers spurn it.
( Appleâ(TM)s SANE got it right for 680x0-based Macs, but lost it upon switching to Power-Macs.)"