Just finished converting my home's systems from Debian to OpenServer. Cost me a bundle, but I figure it's important to support organizations that defend IP rights.
Actually, Apple's pdf tools are cutting into Acrobat's position. For instance, MacOSX users can print to PDF, no problem. Acrobat Reader is still superior in some ways to MacOSX's Preview (text selection, table of contents, find), but in 10.3, Apple's PDF program will supposedly be much improved.
On the other hand, Adobe is planning to release a new version of PhotoShop for the Mac fairly soon, and InDesign is still available (although, if this most recent sales tactic is any indication, they might still run kicking and screaming from Quark).
The earlier study was of polished code, many iterations after release. This latest study is of an unpolished developers snapshot. I suppose that you might be able to divine some kind of wisdom about the development of open-source software-- Development branches shall be as stable as commercial code. Release branches shall be more so.
The metrics report does mention the version number (dev-1/31/03), though the fact that this is development code is not explicitly noted No mentions is made who commissioned this study. Perhaps the company is simply fishing for clients.
If you download the defect report (available from here*, it will explain exactly where the bugs are. For instance, the first bug is
DEFECT CLASS: Null Pointer Dereference DEFECT ID 1 LOCATION: httpd-2.1/modules/aaa/mod_auth_basic.c:291 DESCRIPTION The local pointer variable current_provider, declared on line 235, and assigned on line 257, may be NULL where it is dereferenced on line 291. PRECONDITIONS The conditional expression (res) on line 253 evaluates to false AND The conditional expression (!current_provider) on line 264 evaluates to true AND The conditional expression (!provider || !provider->check_password) on line 268 evaluates to false AND The conditional expression (auth_result != AUTH_USER_NOT_FOUND) on line 282 evaluates to false AND The conditional expression (!conf->providers) on line 287 evaluates to false.
Each bug report is followed by the snippet of source code containing the defect.
The metric report simply reports the statistics. For instance, the most bug ridden file is otherchild.c. The most common bug class is "dereferencing a NULL pointer".
If the Apache developers simply want to fix the bugs, they can use the Defect Report. If they want conduct a brutal purge of their contributors, they can use the Metric report.
*Yes, Reasoning wants an email address. They will mail you a URL (a rather simple one at that) to access the reports.
Gator and Uhaul "have argued that their ad-sales and delivery tactics are legal because consumers agree to receive the ads when they download and install their software."
It would seem to me that adjusting a hosts file is even more on affirmative decision on the part of the user. Of course, when you run your own ad server, slashdot doesn't generate an ad impression. Theoretically, some of these services might allow a ad impression, but subsequently overlay that ad with another. Slashdot still gets paid, but the correct advertisement never shows up.
The latter system could be construed (by a sufficiently aggrieved advertiser) as fraud. Eventually, though, rates for banner ads will decline even more, depriving advertiser supported sites of much needed revenue. Presumably, that's why the Washington Post sued Gator (and won a preliminary injunction last year-- from the same court, to boot)
and yet he didn't use the Intel compiler (a free download) on the P4
The free version of Intel Fortran comes with this caveat
You will not receive the timely response to support issues through Intel Premier Support that commercial customers receive.
Sometimes, timely response to compiler issues is invaluable to a scientific programmer. If a compiler produces a slight inaccurate result in a game, the programmer might not care very much. The same inaccuracy in a scientific simulation could mean that a year's worth of runs will have to be thrown out.
It's still useful (although marginally so) to have a statistic which allows you to compare a 3.4 Ghz Pentium IV with a 2.16 Ghz G5. In theory, you could multiply the clockspeed by the "efficiency" and get a rough idea of which machine will run your code faster, without having to redo the benchmarks every time the manufacturers release a new PC.
I think the operative questions are: Is it easy to optimize for Altivec? Is it easy to optimize for SSE2?
Suppose you have a year to complete a series of simulation runs. After spending a few months writing code, you discover the intel machine, using code compiled by icc is 20% faster than a G5 coupled with gcc.
Now, you could start your runs. Or you could spend another month or so hand vectorizing code for a tenfold increase in speed (at least on the altivec platform). Is it as easy to produce such speed gains by hand optimizing SSE2 code?
My source (http://www.cockeyed.com/science/gallon/liquid.htm l) apparently used.125 oz servings, for a total of 1024 doses. Definitely not pure.
Let's see. 1 floz = 29.5 ml. That would imply a dosage of 3.68 ml. A lot of LSD.
According to the LSD faq when LSD was legal, it was available in 1 ml ampules, each containing 100 micrograms of LSD dissolved in water. Assuming (ha ha ha) that street LSD is prepared to the same standards as Delysid, the $200 ampules would contain 300-400 micrograms. A heavy dose, to be sure.
Printer Ink retails for $2,701.52 per gallon, but that's a mere pitance, compared to "Chanel No. 5" perfume ($25,600 per gallon) or LSD ($122,880 per gallon)
Simple. Recall that there 252 pennies to the guinea. A half guinea is worth 10 and six, or about 11/20 of a pound. 5 newpence (p) are equal to 12 pennies (d). It's all frighteningly easy, once you get the hang of it.
An email button? That's nothing. My keyboard has a "shopping" button. (I think I have it set to open up freshmeat, but even so, I never use that feature.) Plus it has a scroll wheel, fast forward, rewind, play, and stop controls. It's not quite as rococo as those new Microsoft "Office" keyboards, but every time I sit down at my desk, I feel like I'm really using one of these Vroom, Vroom !
He did say it was "snappier". As the owner of a dual-USB iBook, I can say that I'm relieved by that statement.
As a caveat, let me point out that dual-USB iBooks used 8 MB Rage 128Pro (read "obsolete") graphics . As that openGL implementation is only able to handle texture objects with dimensions of size 2^n, early iBooks cannot run QuartzExtreme. Nor can we use fancy T&L acceleration nor even (sniff) shaders.
Though the current iBooks are equipped with ATI Radeon 7500 chipsets, and thus can use Quartz Extreme, shaders are not (IIRC) supported. This omission may not seem all that important now, but in later versions of OS-X, Apple might see fit to leverage shaders for another "Oh my fucking god. I need to buy 10.4" type moment.
The iBooks haven't been updated for quite a while, only support Airport (not Airport Extreme), can't use Altivec, are are generally getting a bit long in the tooth. If I had money burning a hole in pocket, I'd wait to see if Apple comes out with anything new in the next two or three months, or go with a 12 inch Powerbook G4.
"Today we slammed the door shut on (this copyright question) and threw away the key once and for all, so this issue doesn't come up again," said Darl McBride, SCO's president and chief executive officer, in an interview.
That's from 10 June 2003, although things may have changed yet again. To my knowledge, Novell still claims that it owns the Unix patent portfolio.
It is true that some test kernels are unstable. However, the increased stability of release kernels is primarily a result of people testing pre-release kernels. If a video card manufacturer complicates this process, and as a result, the "number of eyes" diminishes, the officially stable kernels will be less so.
Slashdot Mirror
I can't figure out how to download his dissertation. I want to judge for myself whether "tedious and unimportant" is an apt description.
Just finished converting my home's systems from Debian to OpenServer. Cost me a bundle, but I figure it's important to support organizations that defend IP rights.
Actually, Apple's pdf tools are cutting into Acrobat's position. For instance, MacOSX users can print to PDF, no problem. Acrobat Reader is still superior in some ways to MacOSX's Preview (text selection, table of contents, find), but in 10.3, Apple's PDF program will supposedly be much improved.
On the other hand, Adobe is planning to release a new version of PhotoShop for the Mac fairly soon, and InDesign is still available (although, if this most recent sales tactic is any indication, they might still run kicking and screaming from Quark).
The earlier study was of polished code, many iterations after release. This latest study is of an unpolished developers snapshot. I suppose that you might be able to divine some kind of wisdom about the development of open-source software-- Development branches shall be as stable as commercial code. Release branches shall be more so.
The metrics report does mention the version number (dev-1/31/03), though the fact that this is development code is not explicitly noted No mentions is made who commissioned this study. Perhaps the company is simply fishing for clients.
For instance, the first bug is
Each bug report is followed by the snippet of source code containing the defect.
The metric report simply reports the statistics. For instance, the most bug ridden file is otherchild.c. The most common bug class is "dereferencing a NULL pointer".
If the Apache developers simply want to fix the bugs, they can use the Defect Report. If they want conduct a brutal purge of their contributors, they can use the Metric report.
*Yes, Reasoning wants an email address. They will mail you a URL (a rather simple one at that) to access the reports.
arrgh. Should be Gator and whenu.com....
btw, why is Wells Fargo suing both Gator and WhenU.com?
Gator and Uhaul "have argued that their ad-sales and delivery tactics are legal because consumers agree to receive the ads when they download and install their software."
It would seem to me that adjusting a hosts file is even more on affirmative decision on the part of the user. Of course, when you run your own ad server, slashdot doesn't generate an ad impression. Theoretically, some of these services might allow a ad impression, but subsequently overlay that ad with another. Slashdot still gets paid, but the correct advertisement never shows up.
The latter system could be construed (by a sufficiently aggrieved advertiser) as fraud. Eventually, though, rates for banner ads will decline even more, depriving advertiser supported sites of much needed revenue. Presumably, that's why the Washington Post sued Gator (and won a preliminary injunction last year-- from the same court, to boot)
The free version of Intel Fortran comes with this caveat
Sometimes, timely response to compiler issues is invaluable to a
scientific programmer. If a compiler produces a slight inaccurate result
in a game, the programmer might not care very much. The same inaccuracy
in a scientific simulation could mean that a year's worth of runs will
have to be thrown out.
It's still useful (although marginally so) to have a statistic which allows you to compare a 3.4 Ghz Pentium IV with a 2.16 Ghz G5. In theory, you could multiply the clockspeed by the "efficiency" and get a rough idea of which machine will run your code faster, without having to redo the benchmarks every time the manufacturers release a new PC.
I think the operative questions are:
Is it easy to optimize for Altivec?
Is it easy to optimize for SSE2?
Suppose you have a year to complete a series of simulation runs. After spending a few months writing code, you discover the intel machine, using code compiled by icc is 20% faster than a G5 coupled with gcc.
Now, you could start your runs. Or you could spend another month or so hand vectorizing code for a tenfold increase in speed (at least on the altivec platform). Is it as easy to produce such speed gains by hand optimizing SSE2 code?
The price I gave was for many dozens of cartridges, bought at retail.
My source (http://www.cockeyed.com/science/gallon/liquid.htm l) apparently used .125 oz servings, for a total of 1024 doses. Definitely not pure.
Let's see. 1 floz = 29.5 ml. That would imply a dosage of 3.68 ml. A lot of LSD.
According to the LSD faq when LSD was legal, it was available in 1 ml ampules, each containing 100 micrograms of LSD dissolved in water. Assuming (ha ha ha) that street LSD is prepared to the same standards as Delysid, the $200 ampules would contain 300-400 micrograms. A heavy dose, to be sure.
source of the above information.
The natural order of things is listed here
Printer Ink retails for $2,701.52 per gallon, but that's a mere pitance, compared to "Chanel No. 5" perfume ($25,600 per gallon) or LSD ($122,880 per gallon)
did I say 11/20 of a pound? I meant 21/40.
Simple. Recall that there 252 pennies to the guinea. A half guinea is worth 10 and six, or about 11/20 of a pound. 5 newpence (p) are equal to 12 pennies (d). It's all frighteningly easy, once you get the hang of it.
An email button? That's nothing. My keyboard has a "shopping" button. (I think I have it set to open up freshmeat, but even so, I never use that feature.) Plus it has a scroll wheel, fast forward, rewind, play, and stop controls. It's not quite as rococo as those new Microsoft "Office" keyboards, but every time I sit down at my desk, I feel like I'm really using one of these Vroom, Vroom !
XT style? The first XT keyboards had 88 keys, and connected with a Din-5 that doesn't work with modern machines
He did say it was "snappier". As the owner of a dual-USB iBook, I can say that I'm relieved by that statement.
As a caveat, let me point out that dual-USB iBooks used 8 MB Rage 128Pro (read "obsolete") graphics . As that openGL implementation is only able to handle texture objects with dimensions of size 2^n, early iBooks cannot run QuartzExtreme. Nor can we use fancy T&L acceleration nor even (sniff) shaders.
Though the current iBooks are equipped with ATI Radeon 7500 chipsets, and thus can use Quartz Extreme, shaders are not (IIRC) supported. This omission may not seem all that important now, but in later versions of OS-X, Apple might see fit to leverage shaders for another "Oh my fucking god. I need to buy 10.4" type moment.
The iBooks haven't been updated for quite a while, only support Airport (not Airport Extreme), can't use Altivec, are are generally getting a bit long in the tooth. If I had money burning a hole in pocket, I'd wait to see if Apple comes out with anything new in the next two or three months, or go with a 12 inch Powerbook G4.
As usual, McBride is gleeful.
That's from 10 June 2003, although things may have changed yet again. To my knowledge, Novell still claims that it owns the Unix patent portfolio.
That article is from early June. Novell may have backed off since then.
It is true that some test kernels are unstable. However, the increased stability of release kernels is primarily a result of people testing pre-release kernels. If a video card manufacturer complicates this process, and as a result, the "number of eyes" diminishes, the officially stable kernels will be less so.
You'd root for the case to be stuck in Chancery court for a number of decades. Perhaps an appropriate fora would be India?
The text merely accompanies the photos.
Pure math is math unsullied by notions of practicality.