Sinful SMS, Shaggy Solaris
Letters A sackful of interesting mail - on Australia's broadband woes, and the great P.Eng debate to follow. But here's a fascinating miscellany on some recent stories.
Geeks bearing gifts - dodging taxes too?
to fight AIDS in India. The fact that the donation came as part of a coordinated effort to dissuade the Indian government from its plan to adopt Linux gave me some qualms, but I was still happy to see that much money directed to fight the plague that threatens to cripple India.
Today I was extremely disappointed to learn that it wasn't Bill Gates who donated the money, nor was it Microsoft, but it was the Bill and Melinda Gates
Foundation. I had thought that the B&MG Foundation was a charity whose Board of
Directors would decide how to spend the Foundation's money in the interests of global health and education, not an extra, tax-free, $24 billion war-chest for Microsoft to use in strategic maneuvers.
The Bill and Melinda Gates Foundation is the world's richest 501(c)3 charity by a very large margin, holding assets worth $24 billion. I wonder how much tax would be due if it was decided that instead of being an independent non-profit
charity, it was actually a subsidiary of Microsoft, Inc.?
I hasten to add that I do not wish that this would happen!
The B&MG Foundation is uniquely empowered to save billions of people from disease and ignorance in the coming decades. It has already donated over 5 billion dollars for valuable philanthropic projects, focusing primarily on global health and education. It would be a terrible tragedy if the impact of their work were diminished by the perception that they serve two masters: the greater good for all of humanity, and the continued dominance of Microsoft's monopoly over all of humanity's computation.
Thank you Zooko. Let us banish the cynical thought that this is a gigantic tax dodge once and forever.
That said, isn't one human life made less miserable a good thing?
Dirty Unix Secrets
Two terrific letters with observations on modern Unix. One we quoted in part when Sun's Jonathan Schwartz floated the idea that Solaris might go LSB compliant. Here it is in full.
"Open" Solaris being LSB compliant? I shake my head at this concept.
This is like taking a pure breed dog and intentionally cross breeding with a mutt. Solaris has a solid SysV heritage, and SunOS was as true to BSD as anything. Why they would mate it to a demonic half breed like Linux is beyond me. This is coming from someone who hates SysV and all it stands for.
I am happiest running FreeBSD on servers and my personal desktop but my second choice is Solaris.
Why? Because every thing is where it is supposed to be, things like system calls and arguments don't randomly change like oh say, Linux. Sure I have to put up with the fact that things like /dev/random come in a seperate package/cd on solaris, but all and all its stable and standardized as per SysV. If Sun starts installing everything under the sun in /usr/bin like linux then I'd rather have AIX or a hole in my head.
God help us all if Solaris goes LSB compliant.
I didn't fight like hell with Sun to bring back Solaris x86 for this kind of garbage. Going LSB compliant looks good for the linux mob, but it's going to irritate the hell out of entrenched Solaris crowd.
An even more entertaining broadside takes no prisoners, but is a useful antidote if you're fed up with hearing OS X described as "clean and modern"…
I just finished evaluating JFS 1.0.24 for Linux. My opinion of 1.0.24 and JFS is IBM is doing the port as a courtesy to AIX and OS/2 migrators. It is extremely robust, but slow, 2x slower than XFS or Reiser. 1.1 XFS for Linux is unreal. It does "everything," it has done it for years, it's high performance, has a robust heritage and is very good all around.
EXT3 is a dirty hack (EXT2 with fake journaling). Reiser is a UFO, and is easily corruptible, and I fail to understand its wide use and early integration in the kernel - my only guess is its simplicity required the least cleaning up of the kludged Linux file system underpinnings.
UFS+logging on Solaris and UFS+S on FreeBSD are both superior, as well as Veritas (available on NT and Solaris).
My experience with Mac OS 10.X, including 10.2 has been horrible. I think its inferior, the Mach kernel was deprecated by its progenitors, CMU, in 1994. I think the FreeBSD userland is outdated. I think HFS+ is a pathetic file system and fail to understand why they don't use UFS,
[Application compatibility - ed.] But if you have ever tried using it with OS X you know its not "finished."
Adding journaling to HFS+ will only slow down an already horrifically bloated and underpowered platform. I find it laughable Apple hardware does not get submitted to www.spec.org, but I have CPU2000 results for PPC 1.25GHz, and of course it is so horrible they can't submit - everything including the SPARC beats it hands down.
It is unfortunate that OS X exists. It distracts people from a real endeavor like FreeBSD 5. Most Mac zealots piss and whine and pine for the good old days of OS 9 anyway.
(I do not deprecate the use of OS X if you like it. I personally would refuse any other rendition of MacOS.)
But I wanted FreeBSD 4.x userland, and a real kernel. Jobs' fetish for mach and NeXT is disrupting a would-be industry killer. It pains me to use Office X and IE as the default browser.
"BSD is for people who like UNIX. Linux is for people who hate Microsoft"
Which goes to prove that the "best of both worlds" UNIX plus UI can for some be the "worst of both worlds".
Jorn Barger has been doing some terrific archaeology on UNIX on early micros here. His GNU/Linux timeline also unearths some Usenet gems - Linus skewering crufty sig files is worth the price of admission alone.
Subject: re: SMS is a sin story
Despite its official sounding name the National Human Rights Council has nothing to do with India's (official) National Human Rights Commission. They specialise in colourful "protests" on various topics.
Recent "protests" have included one against the rain (or lack of it) in Delhi's hot summer - they organised a classical Indian dance performance, on the street, in front of parliament, to appease the rain gods. Earlier, they had people marching wearing nothing but loincloths (like traditional Indian wrestlers) to protest against the popularity of chewing tobacco. protesting against SMS is more high-tech, though.
Rishab Aiyer Ghosh
PowerPoint Rangers, Corridor Warriors
I assume you've heard of the "Powerpoint Ranger", those Pentagon bureacrats that are deadly in the use of Powerpoint in making nice presentations to fight terror.
Here's the slideshow of the PowerPoint Ranger creed.
You missed my favorite office title, Chairborne Ranger.
Microsoft did NOT originate the phrase "corridor warrior". I do not know for sure who did, but for reasons unknown and likely having to do with near terminal boredom, I scribbled the phrase, and some rather rude comments regarding the person who said it, on my notes during a meeting when a Tandy rep was attempting to sell the firm for which I worked their very first portable. The notes are still in my possession. I may not recall the model correctly, but I think it was "100".
RIM's magnificent keyboard invention
Handspring settles Lawsuits in Motion suit
Geez Andrew, take a pill.
RIM was advised by its lawyers that certain competitors were infringing
it's patents, so they asked them to stop. What in Hell is wrong with that?
What's the point of going to the expense of obtaining a patent if you're not
going to enforce it for Chrisake? [to barter some IP that you haven't got - Letters Ed.]
Are you equally miffed at Texas Instruments for enforcing it's patents on the "hand-held calculator"? They collected 100's of MegaBucks on that one at least 15 years after those devices had become $3.99 check-out line pickup items. AND GOOD FOR THEM!!
If you think that this, or any other patent is frivellous, vent your anger where it belongs, on the US Patent Office. These are the guys that issued a patent to some slime ball from Arizona on a strain of beans that Mexican farmers had been growing for centuries so that he could charge them $100/truckload (more than their cost of production) to bring them into the States. These are the guys that should be shot and pissed on, not innovative companies like RIM and TI (and Handspring, and Nokkia and others, for that matter).
Stoney Creek, ON
"Obviously something quite significant is happening, and it isn't happening here."
Technologically, undoubtedly. But your piece about RIM shows that innovation is alive and well in the field of aggressively trivial litigation - RIM are great "litivators".
And I think that's what MS were really fighting for: their right to litivate.
Actually Duncan, Microsoft is one of the least aggressive litigants when it comes to its IP.
Pet Store Boys
Microsoft and possibly the authors of the study Still Don't Get It.
What was the Cost of retooling the Pet project?
It looks like at least 90 days of work by at least 4-8 highly experienced programmers. $50 to $100 dollars an hour for 8 programmers: $1600 per day * 90 days:
What's the cost of simply moving off an Intel generic platform and onto a
Sun server? $20,000-40,000?
Java optimizes programmer expense. Java optimizes programmer productivity as witnessed by the number of code lines. Also, Microsoft's solution still uses Com+ components. So, it's not truly .Net architecture.
There are a number of key areas that where not explained correctly so the report is largely useless...
1. JDBC drivers can play a leading part on the performance of load tests. I have tested numerous JDBC drivers and find the Oracle drivers to be quite
2. Why was not the same database used through out the tests. Its obvious that
.Net is going to perform well with MS SQL Server since it is basically tied
in with it (It is a .Net component). Try changing .Net to run with Oracle...what different would that make?
3. Windows was chosen over Linux because it was found that J2EE application servers performed better on Windows. What was there setup criteria for this conclusion?
a.) Did they run Linux as standalone fronted by IIS on a windows machine or did they setup clustered Apache Web servers on Linux with clustered Linux
Applicaton servers behind??
b.) Did they slim down the Linux kernel so that it was only setup to run the application server services.
4. Partly linked with point (3). Application server clustering...What happened to this? It does mention clustering (in two places) but they do not mention how many application servers existing in the cluster, if a cluster existed at all.
5. Cost of ownership? JBoss is free which can run on Linux which is free which can also connect to Postresql/MySQL database servers, which are free and also can run on Linux.
Total software cost = ZERO
My conclusion here is that if we had a proper test environment setup inviting each company to come in and tune their own application servers and also choose what operating system and database is used (as long as they keep to the same hardware spec!) Then I can guarantee that we will see more of a less biased result...what I would expect is with a single high spec Intel architecture machine that .Net would probably just pip Java but in a clustered environment using linux on the same spec machines linked in with
apache up front then I think we would see a different picture...
Whatever Microsoft wins from their benchmark "win" is completely undermined by their licensing language.
Anyone who has fixed a bug in any Microsoft product in the past year has, as part of the bug fix, promised not to talk about dot net benchmarks without clearing the answers first with Microsoft. Since everyone needs to patch some supremely dangerous holes in IE or XP, or to upgrade media player to play this week's media format, we have ALL agreed not to criticise dot net's performance unless Microsoft gives permission.
If dot net was as good as Java, then why stuff "silence" clauses down the throat of everyone who uses unrelated Microsoft products?
I can only assume that the bizarre clause was inserted into the licensing agreement for a reason
J2EE is a bundle of overhyped, oversold technology architecture based on the
fallacy of write once run anywhere.
I took a look at J2EE about 2 years ago - wasted about 2 weeks - and quickly moved on to Jini - now that is a piece of Sun technology that is underhyped, undersold, does *everything* J2EE does in a fraction of the time and cost - whose concepts are modern, and clean.
What I don't understand is why other people just don't "get it" ? Any ideas?