Microsoft had more than 1000 people working on Internet Explorer alone.
The 10-20 LOC/day rate also includes a lot meetings and excessive documentation before, during and after coding, and endless debugging sessions.
Working 40 hours per week almost alone in a two-people office without phone on a familiar task with guru level programming language skills some people can churn out complex tested high quality code at a sustained rate of few hundred lines per day.
On the other hand, if you are sitting with fifty other people in one room with frequent interruptions, several six hour meetings per week, coordinating with some hundred foreign programmers in different timezones around the world trying to implementing something specified in this six-thousand page document nobody really understands, working under high pressure twelve hours every day including weekends. you will easily find your output below 10 LOC/day.
During the last 20 years I have probably spent more than half of my awake time staring into a computer or TV-Screen.
Sometimes, especially in spring, when I walk around in the woods and watch brown and yellow leafs, sometimes some colors feel strange, like an itching inside my brain. Once I stared for several minutes into some square meters of ground covered with yellow leafs reflecting the direct light of the sun. I was happy and excited just by looking at this color, which in fact was not like any yellow I have seen before in my life. It seemed to be yellow, orange, brown and green at the same time. The sensation felt great, and I after a while looking at it I was sure this color could never be reproduced on an RGB display. Sometimes I also feel slightly disoriented when bathed in red or blue light from flourescent lamps, which are placed by artists in some public places here, and I like this sensation of brain-tickling so I often to walk through there just to get this strange feeling.
Regarding MP3, when I started to listen to 128kbit MP3, I was hardly able to recognize a difference. Later I became very sensitive to the artfacts of MP3-compression and almost stopped to listen to 128kBit MP3 for some types of music, especially Madonna and The Beatles sound really bad to me when compressed at 128kbit. I could not hear artifacts with 128 kBit OggVorbis Compression, but I am curious whether I will after a while. Also this effect of feeling bad while listening to an MP3 Versions seems to happen with music I heard very often ucompressed before.
The article does not give any answers, but it raises some really important questions. However, I am afraid they will never be answered, because after more than fifty years of television the consequences are still not understood very well.
,..one can ask what the return has been. [...]...scientific projects...space research
The wrong idea that sending people into space is science and research goes all back to the Apollo program, when after the first landing on the moon NASA tried to sell subsequent missions as "scientific missions".
IMHO, sending a man to the moon was the highest cultural achievement of mankind in history so far, but as a piece of art, there is no much value in repeating it, and as nobody had the balls to admit that hundred billion dollars were spent for art, it had to be science.
There is plenty of important science that happens in space, but you don't need people hanging around for that.
Manned Space Exploration is about beeing there, and to feel how it feels to be there. It is about living there. It is about building houses, planting trees and fathering children out there. And cruising around with a cool car, if you are american.
After Apollo 17 all space programs world started to decline, and there is no end in sight. The Space Shuttle program started by crippling Wernherr von Braun's original design that had a piloted, horizontal landing reuasable first stage by using a cheap throw-away fuel tank and reusable solid fuel boosters, ending up with a Space Shuttle with more expensive payload than using throw-away rockets. The buerocrats way of wasting money by budget cutting. And every news I heard about the ISS the last twenty years was about budget cuts and delays. I heard you need 2.5 people just to operate it, and there are three guys up there. SNAFU.
It is sad, and I hope I will be wrong, but within a decade we will see:
The ISS will be abandoned a finally reenter the atmosphere
The last Space Shuttle will go out of service
There will be no more capabilities to send humans into orbit any more
I just hope mankind will regain manned space travel before we will have depleted our natural resources here down on earth.
To find out what a patent is about and who will be affected, you have to read the first claim, find possibly other independet claims and ask yourself what kind of system will have all of these properties mentioned in one of these independent claims.
In this particular case you have just four criterias in claim 1, and the are pretty unspecific, so it it is a patent possibly dangerous to many people. There are two additional independent claims 4 and 7, which you can view as different additional claims that were put into the patent to widen it's scope. The rest of the patent just clarifies and specializes these independent claims.
It is the examiner's job to narrow the claims as much as possible, and the applicant usually want to have them as wide as possible. Here, definitely the applicants did a better job than the examiner.
From what I see, there is no real invention here, but that is true for most of the so called IT-Patents, and this one is not a particularly bad example, it is merely a typical patent you often have to write because the competition does it too.
Thank you for the hints on thread local storage; I am glad to see this has been adressed.
Regarding changing of priorities, I think that with SCHED_OTHER the priority is beeing automatically modified by the scheduler to distibute cycles in a more fair fashion.
I tried both SCHED_RR and SCHED_FIFO and changing priorities basically works, but it seemed to me that changing priorities did not have an immediate effect as required to implement priority ceiling locks.
For example, when boosting the priority of a thread to the ceiling priority, and the thread is the only one with this priority, I expect it to run without beeing preempted by anyone before the priority is lowered or the process blocks. On the other hand, when lowering the priority, I expect a higher prio thread to be executed immediately. I would also expect the order of unblocking threads is correctly adjusted when their priority was changed while suspended.
However, it seems that priority changes do not much affect the actual timeslice or the unblocking order, but I did not have the means to find out what exactly happens; using a debugger is outright impossible with fine-grained multi threaded programs.
Is it possible that some system thread needs to run inbetween to do some housekeeping ? Do you have any hints about the scheduler's inner workings ?
Among the issues with threads beeing half a process, half a thread (getpid() bug, signal handling etc.) that are mentioned in the article, I found issues in two other areas:
scheduler does not immediately respond to priority changes
thread-specific storage access is slow
There is a well known effect in multi-threaded programming called priority inversion that can cause deadlocks when a low-priority thread has acquired a resource that a high priority thread is waiting for, but a medium priority thread keeps the low priority thread from beeing executed and so the medium priority thread effectively gets more cycles than the high priority thread.
One way to overcome this problem is to use priority ceiling locks where the priority of a thread is boosted to a ceiling value when it acquires a lock. Unfortunately I found that changing the priority of a thread for a short interval does not have any effect at all with the current 2.4.x standard pthreads implementation.
The second problem I ecountered is that accessing thread-specific storage with pthread_getspecific() takes 100-200 processor cycles on 1 Ghz PIII, which makes this common approach to overcome hotspots almost as slow as locking.
Does anyone know if any of these issues are adressed by the new implementations ?
I made exactly the opposite experience; after switching to XFS, we had no problems with corrupted filesystems any more, on about 30 different development machines.
I will never use ReiserFS again because this trouble I had.
From the engineering point of view, you have to drop some legacy from time to time, or either your progress will slow down, or you need more horsepower to keep your speed, like the Microsoft IE development with more than 1000 people, or like the Windows XP dev team beeing 10-30 times larger than the Windows NT team.
And it is perfectly in the tradition of Apple, in contrast with the Microsoft/Intel tradition of keeping A20 gates and VGA modes more than ten years after nobody really uses them any more.
Apple would either require a new OS 9 release for every new Mac, or had to keep the current system design, unable to benefit from new processor or bus technologies.
I have been a long time Mac User and developer between '85 and '99, but Apple lost me to Windows and Linux when fucked up their new OS development (Rhapsody etc.). I would give them a try when the gcc compiler on their top-of-the-line laptops would be at least 50% as fast as on a current Dell top-of-the-line system under Linux. (It was five times slower the last time I tried.)
I participated in the test myself, and it was a true blind test; it was really hard to tell between the three best sounding files at 128kbit, but there were definetely three codecs better than the rest at 128 kbit. And after receiving the results, I was glad that I could trust my ears, as I correctly identified the uncompressed version at 64 kbit and put it 2nd at 128 kbit.
I am really annoyed by MP3 artifacts, and I observed that I became more sensitive to them during the years of listening MP3s. This coincides with blindly placing MP3 on the bottom in both tests.
I am really excited about the outcome and it will have two consequences for me:
1)I will get an OGG Encoder/Player asap and encode my CDs in OGG from now on.
2) I will enjoy listening to compressed music much more than to date as I know now that I wasn't able to tell the difference from the original at 128k.
Here are my personal test result, and at 128k I was already a bit tired, and I could not really find a difference between the top 3:
Ihre Bewertung für 64 kBit/s-Codecs.:
Platz 1: unkomprimiert (WAV) Platz 2: AAC Platz 3: Ogg Vorbis Platz 4: MP3Pro Platz 5: RealAudio Platz 6: Windows Media Audio Platz 7: MP3
Ihre Bewertung für 128 kBit/s-Codecs:
Platz 1: Windows Media Audio Platz 2: unkomprimiert (WAV) Platz 3: Ogg Vorbis Platz 4: AAC Platz 5: RealAudio Platz 6: MP3 Platz 7: MP3Pro
On SIGGRAPH last month I was told by ATI Engineers that the new Linux Drivers from ATI will not be Open Source; they will follow NVIDIA in this. I have no experiece with the current open source drivers, but six months ago there was *no* hardware T&L support a.k.a. 3D hw accelleration with these drivers.
1) Try to make sure that your overtime does not come for free, or you will be doomed. Ask for compensation in free time after the project will be over, and make a written agreement about the terms, exact dates. Propose a reasonable cutoff like 2 hours per day and 8 hours on weekends. Do not agree on additional money.
2) The best time to talk about this is when start a job; talk about overtime policies explicitly, how much is excpected during normal times, during a crisis, and how much you are willing to give under these circumstances.
3)Buy, read and cite "Peopleware" and "The Deadline" to your managers; "Extreme Programming" even goes that far to have a "No Overtime" policy as a part of the process rules.
4) As as Team Leader, you have to fight for your Team and protect it from impossible demands from upper management, even risking to be fired.
5) Divide the Project into 20 to 50 tasks, and let everone on the team estimate the required time for every task; present the anonymized results to the management.
6) When there is crisis, productivity is not as importing as successfully meeting the deadline; every person on the team has to find individually the point where the most work is beeing done per time, while "Work Done" means 100% completed functionality meeting the required defect rates, not 90% ready bug ridden pieces. Depending on age, type of work, and current personal condition this optimal point lies between working 6-11 hours daily, at 5-6 day a week.
7) Typical Management often is not as mentally exhausting like development; Large meetings, travelling, reading and writing messages can feel like spare time compared to complicated design, coding or debugging sessions.
8) Try to avoid getting into thinking "them" against "me/us". Offer every support and idea to meet the business objectives.
9) You have the right on a chance to be successful. Make yourself aware and explain to your management that they are denying you this right, and take consequences if it does not change: Either quit or pick and stick to an elaborated "doomed project survival strategy".
10) The is no such thing as a free lunch. Overtime has many associated cost; quality, productivity, health, well-being at work; make sure that the bill goes the one who ordered.
"A mother needs nine months to give birth to a child, and even four mothers can't make it in five months. And contrary to the belief of some managers, even twenty mothers can't do it faster." [Kent Beck, IIRC]
Java was supposed to be a small, open platform for writing clients that are delivered through the web browser.
Java was never supposed to run in a Browser or the Internet at all; see a short history here. It was started 1991 without any special purpose in mind and was then focussed 1993-1994 to run on these "Interactive Television" set top boxes. After this failed there were some plans to use it as CD-ROM/online multimedia platform. Finally two guys wrote a Browser in Java (remember HotJava?), and only in 1995 Netscape licensed it for the Browser.
IMO Java is far from dead anywhere and is here to stay, as many mediocre technologies that were present at the right time with a successfully balanced set of tradeoffs. My language of choice is C++ and programming Java is like typing with one Hand in my pocket and half of my brain asleep, but it is the best choice for a beginner's language.
C# has no justification between Java and C++, and as a Visual Basic substitute it might succeed only in the long run.
Microsoft does not have the power to kill a programming language or guarantee it's success, as the end user does not care about what programming language something is written in, and the programmer will pick and stay with the one that he feels comfortable with.
Jim Independent, a WindowsWorld.com columnist, compares Linux and Windows installations. He expected Windows to be faster and easier since Microsoft has been at it for 21 years. (DOS 1.0 was released 21 years ago today.) It turns out, Windows is quicker and less manually intensive."
Honestly, it is easy possible to find configurations where I can prove either view. In general, it is still a pain to get all hardware supported and configuered under linux; wlan, firewire, cameras and high-end audio devices are just a few examples I usually spend days to make them work properly.
In early times, you had to make your own PCB design for your machine, solder it together, write your own kernel, write some Libraries and and put together almost any app within a reasonable time.
In the late eigthies, you still had to write almost every application if you required something like a workflow and wanted the system to be used by mere mortals.
Today, one person alone can set up an integrated office system with document management, customer database, accounting, reporting, E-mail access, firewalls, printing, backup, telephone integration, banking, video and audio editing, ftp and webserver, dns, dhcp, wavelan, gigabit backbone, raids, computing clusters and mobile vpn access within weeks, and several hundredthousand apps are available for free or very little money. You can easyly afford a database software or an office-suite worth a few hundred million dollars of develepment effort, and get even millions of man
-months delivered on a $50 Linux Distribution, including the source code. Or if you buy a Mac or a PC today, you can actually do useful things without even installing additional software - remember, what you could do with an AppleII out of the Box ?
So today, if you want to roll your own stuff, or even spend a few days on improving your customer database access, you need many thousand customers to justify even using a real database instead of MS Outlook contacts or a simple spreadsheet.
Trying to build a new 3D-Engine, a Web-Browser, a database engine or a new GUI library is almost insane from a business point of view, so the deeper you descend into the swamps of IT-Development, the better your justification has to be for shelving out that money and taking the risk of failure.
I try to ride Moore's law and aim for something unique that does not exist so far because it was impossible or too hard to do in the past. I try to stay current by spending a lot of time hands on new technology, and I steadyly improve my and my team's knowledge and skills, but I admit: it is increasingly harder to find and exploit those niches where you have both: Fun and Profit.
OTOH, there will always be the Linus way: Build something for the sheer fun and knowledge, and the worst case is that you are happier and smarter afterwards.
And if you don't mind to listen to someone who has been around for a while and covered some distance: Never do things for the only purpose of profit. It will minimize your chances, but even if you succeed, you will not be any happier than today. And probably have fewer friends.
Fine joke, except: There is no *dark* side of the moon; the moon's rotation locks one side to the earth, not to the sun, so the *far* side of the moon gets as much sunlight as the near side.
F16 flipped upside down when crossing the equator
on
Pet Bugs?
·
· Score: 1
...while controlled by the autopilot is my favourite pet bug.
According to http://www.cia.gov/cia/publications/factbook/
the death rate in the U.S. is 8.7 deaths/1,000 population. Out of the cited 20.000 researchers, there should be 174 dead Bio-researchers within one year, or 72 dead in five months.
Or the other way around, 11 dead in five months would correspond to a community of about 3000 experts, which seems to be a probable size.
As you can see, it is all well within the normal death rate. Nothing strange here, just some math-handicapped newspeople making up a story out of nothing.
I saw ironically in a feature on German TV about ten years ago. There they had shown much more severe research results about how watching TV harms your health:
It causes overweight not only because you don't move around, but it slows down metabolism while it makes you feel hungry and lets you eat tons of snacks while not beeing aware of it (->fat)
While your body goes into a sleep-like state, you feel very exhausted after long TV sessions; this really wrecks havoc on your immune system(->sick)
Watching TV before sleep forces your brain's nightly garbage collection to work on what you have seen on TV instead on your real life experiences, with negative effects on decision making and dealing with real life problems, up to causing serious mental diseases(->mad)
Watching TV makes the parts of you brain responsible for controlling your motion wither, which is known to negatively affect your intellectual capabilities (->stupid)
And finally the incredible amount of trash transmitted over U.S. TV can not have any positive effects on the spectator, and all this is regularily interrupted with a mixture of ads, many of them using sophisticated techniques to manipulate your mind. (I can not stand a U.S. TV program for more than a couple of minutes because among other things I am not used to these frequent breaks; it makes me so agressive that I have to switch it off)
I hope that one day the people responsible for this lunacy will have to pay for it.
And I love it. I did not have any trouble with it at all; it worked from the first day, and I never experienced any failures, and I did not take special care; it usually resides in the ipaq jacket when I carry it. The only problem is the power consumption; you should use it only with the PCMCIA-Jacket which has an built in additional LI-Ion battery; I would not recommend using it with the slightly smaller CF-Jacket; it drains your batterie in less than one hour playing MP3s, while with the PCMCIA Jacket it lasts about 2-3 hours, which is acceptable in most cases.
But the best was from the beginning that you simply insert it into your laptop and just drag all your music directly from your terabyte-server onto the disk, or play the stuff on your laptop when the ipaq ran out of power.
The real write rate I measured is between random 1124 and sequential 1260 kbytes/sec; read was beteen random 960 kbyte and sequential 1260 kybtes/sec, access time about 20 ms; It is much faster than anything I have seen with any USB device; the speed has never felt like a problem.
Other unexpected uses were out-of-the-box file transfer between an PowerBook and Wintel-Notebooks, and you can also store enough pr0n on such a disk without taking away much music capacity. You can also carry around a huge library of e-books without caring much about space.
So far I did not regret the investment, and I haven't seen anyone who wasn't impressed when you handed him the smallest one gigabyte harddisk in the world: it feels heavy and expensive in your hand, and my fingers still slightly sweat when they touch it.
p.
P.S.: It still always surprises me how small it is, and the PCMCIA-Adapter is very IBMish: it hides the connector when you remove the drive, and locks the cover to prevent unintended exposure of the contacts. It also comes with watertight two watertight small boxes with rubber polstered corners to safely stow away the PCMCIA-Adpater and the drive.
A small elite minority tries to define a "Body of Knowledge" consistent with their beliefs and slow down progress by carving it in stone and steel, and even seeking political help to limit access to the field for people who refuse to swallow their bullshit. I admit there is some useful stuff in it, and it all sounds very reasonable, but honestly: How many professionals with a master and 4 years of experience in the field (the target audience) are able to professionally apply a substantial portions of this "body" under the correct circumstances ? My guess is less than three out of hundred, and even they will ignore the majority of it because it is useless.
This self appointed elite was not able to set up a website that survives the slashdot effect, and if you ever worked with some of e.g. Rational products and assume that they used their own teaching and tools in constructing this crap, it becomes obvious that no amount of impressive sounding formal engineering methods and a big names lineup alone can produce a good piece of software.
What these methods are good for is to save your ass when the lawyers come, and to prove that you exactly built the shit the customer has ordered, and that it is not your fault that the system is too slow and too hard to use and that the operators should have sticked exactly to the procedures described on page 972 of the operations manual to avoid [two days downtime, loss of 6 months of transaction history, hitting the mountain, reversing the thrust in midair, melting the reactor core]
The attitude that comes out of this document drives me mad, and it has the potential to do so much harm to the evolution of the field that Microsoft seems to be a pleasure compared to these guys. They even try to borrow ethics and principles from medicine, a field where mystification and deception are a guiding principle since 2000 years, and to get a medical education in the roman empire you literally had to become a doctor's slave, being freed not before you were going to replace him.
When I was young I was a firm believer of almost everything my teachers and professors taught me; you may call me a naïve coward, but I was taught many times that it is a bad idea to question their doctrines for several reasons:
1) They are often correct
2) It is difficult to find out when they are wrong
3) If you find out they almost never admit they are wrong
4) They prefer to defend themselves by ridiculing, threatening and insulting
5) They have a lot of power by controlling your career path
There are exceptions, but you need a lot of luck or persistence to encounter one of them; I rarely did, so in a mixture of good faith and cognitive dissonance I spent a lot of time and effort on learning their doctrines.
But soon it became increasingly difficult to bridge the gap between many of those things I learned in academia and the first hand experience I gathered in various positions in the industry, ranging from designing hardware and kernel for embedded factory automation systems in the mid 80's to distributed virtual environments today.
During my professional career I encountered a number of programming languages, operating system, implemented ITU-, IEEE-, DIN-, ISO-, IETF, OMG and W3C standards, and went through some meters of scientific papers, produced at least a half million lines of code myself and guided the production of about five times as much.
And the more I learned from experience, the more livid I became at those teachers. What a waste of time it was, and what a large pile of bullshit I had been forced to swallow, and it took some time to get rid of all those misconceptions that were stuffed in my mind.
Don't take me wrong; I still believe that the scientific way of finding some truth is superior to almost everything else, but a large amount caution is required when dealing with scientific results, and much more caution when dealing with people who are paid for practicing and teaching science.
I am convinced that:
1) most scientific publications are obsolete (about 98% is my guess)
2) there is a tremendous abuse of scientific rituals in areas where science is hardly applicable (e.g. psychology, social sciences, medicine, business, software, history, politics, city planning, art)
3) you can find a bunch of scientist to support almost anything if you throw enough money in
4) real progress in science is rare, comes by surprise and often ruins the life of the contributors
5) most scientist are impostors
Now, coming to the engineers, who are supposed to apply science to build something useful, I am think that:
1) Engineers either think that they are the better scientists or do not know anything about science
2) most engineers are cowards
Now, when it comes to computers and process of creating software, I surmise that after gathering 25 years of experience, learning from failures and steadily improving my skills I feel qualified to raise my voice on this issue.
A simplified, unbalanced summary of my convictions regarding software is:
1) Heavyweight formal methods are a useless waste of time if want to successfully create software that satisfies you and your customer, except in extremely rare cases
2) There is no substitute for own experience, even if you can gain a lot from the experience of others
3) There is no substitute for thorough testing
4) It is helpful to have a solid mathematical knowledge, but you must not confuse math with software
5) Standardization is a necessary evil, and I commiserate those brave souls involved
6) Most standards are flawed because they are too complex to be fully understood even by their creators, not to mention mere mortals
7) Every implementation of a standard I have ever seen was flawed
8) The future belongs to light-weight, easy-to-understand iterative methods
Regarding all those cited mission-critical systems in avionics and telecommunication, most of them are of so incredible bad quality that it is a miracle that mankind survived so far; but even if the system works as designed, people often get killed because the interface between the human operator and the system does not work as expected; just browse through an arbitrarily chosen risks digest, and the first one I just picked (http://www.infowar.com/iwftp/risks/Risks-19/19_94.txt) revealed a perfect example for this: A train derailed because of a bearing overheating that was indicated by an alarm 29 hours before the accident, but the alarm was disconnected because the crew was unable to find the alarm cause and assumed a defect in the monitoring system.
NASA and all other space agencies loose rockets and satellites all the time, several passenger planes were lost because of interface problems, and if you have ever watched an air traffic controller in front of his computerized radar system you will be surprised about the amount of wrong information on his screen which he has to filter out mentally.
Now these areas are where formal methods were applied for many years now, but the effort is extraordinary and the results are not convincing.
I don't need to mention the poor quality of today's popular operating systems (Linux is not better here than Windows). The most popular compilers of one of the most popular programming languages (C++) are so full of bugs that their usefulness is severely affected, but if you try to use development software that is less popular you feel like moving through a mine field: every new move you make might blow it up. It seems to be a general rule that software tends to be as poor as tolerated, no matter how bright the developers are, and only widespread use of the software improves quality.
And those products from the formal methodologist camp are not a nibble better, on the contrary: Rational Rose, for example: After so many years they still can't properly display text, zooming in the diagram views is the worst I have ever seen in any app, and it crashes more often than anything from Microsoft. (And that Rose doesn't understand C++ is no miracle; no program does)
Formal Methods are like communism: In theory, they sound good. In practice, they do not take into account the diversity of human desires and thinking, and the gap between the software engineering community and the people developing software is widening.
I would appreciate this SWEBOK if these guys were not so serious about making life harder for many people and trying to force their "body" down of throat of thousands of students.
They are marginally improving quality in some areas while hampering possible significant progress that comes from widespread use of light weight methods like Extreme Programming, which is the greatest advance in Software Engineering I have seen in my career so far.
In early days the Computer Science was considered as a part of Mathematics, which is plain wrong.
Creating Software in fact is a discipline requiring more artistic than mathematic skills; the best analogy in the "old" world is the Architect who designs houses; he also needs to be familiar with math and needs to know a lot about physical properties of different materials, but the real issue is to create a building where people like to live in,which requires to know more about people and what makes them feel comfortable, more about the history of art than the history of stone and steel.
Creating Software is about making a computer useful to people. Software is not for Computers; they feel equally comfortable with any piece of code; software is about modeling real world entities, and about creating interfaces to the human mind.
Even if you write a protocol stack or a kernel, something no normal user is exposed to, you are creating an interface that makes it possible for another programmer's mind to deal with it.
So it is not the silicon side of software, but the human mind side that makes the difference.
And now throwing in the multi-media aspects of today's most software it becomes obvious that in fifty years creating software will be considered as much an engineering discipline like making a movie today is considered as an engineering discipline because it is an opto-chemical process.
Btw: ever seen somthing funny, beautyful, lovely or cool come out IEEE ?
p.
CD "Protection" possibly a felony in Germany
on
More Copy Protected CDs?
·
· Score: 5, Interesting
"Possibly soon German judges will have to deal with the question whether intentional violation of the Red Book Standard by manipulating the TOC is a criminal use of 'incorrect or incomplete data'. It will be also important wether digitally reading an Audio CD into computer memory is considered as data processing."
The German Criminal Law has in the section "fraud" a the special Paragraph 263a about "computer fraud", which I translate as follows(IANAL, but I am married to one):
263a StGB Computer fraud
(1) Who damages the assets of another person with the intention of providing himself or a third person an illegal pecuniary advantage by the fact that he affects the result of a data processing procedure by incorrect design of the program, by use of incorrect or incomplete data, by unauthorized use of data or otherwise by unauthorized interference with the process is punished with imprisonment up to five years or with fine.
This paragraph is usually used against people tampering with ATMs or a company's computer systems, but I see no reason why my personal computer should not be protected by this law, and for me it is unquestionable that all these CD copy prevention systems are feeding incomplete an incorrect data to my computer, and reading a CD into the memory of my computer is definitely a data processing procedure.
As I know the courts, proving damage is often the crucial point. I think that a clear label stating "This CD is intentionally damaged so that it does not play on most computers and some CD players." would probably save the publisher, but anything short of this IMO constitutes fraud.
In this case there was no warning sign on the CD, so when I buy such a CD I do not get what I pay for. If I bought the CD just to listen to it on my MP3 Player or use a duplicate in my car stereo (my CDs do not last very long when using them in the car), then the value of the CD is zero. And even if I get a refund, it probably takes more than an hour to manage it all, which is a significant amount that easily surmounts the value of the CD: I am tricked, my assets are damaged. This is even simple fraud under 263 StGB, and even a particularly serious case under subsection 3 (repeatedly defrauding a large number of people), which is punished with imprisonement up to ten years.
I am very curious about the outcome of this case, but I would be happy if courts would stop greedy corporations trying to screw me with intentionally damaged products.
If not, I will react like many people: I will stop to buy any music at all; I will fill up a terabyte disk with enough music for the rest of my life in one afternoon, and those record companies can fuck themself.
We are experiencing a lot of stability problem with our applications, and I am very suspicious that they are kernel related. We switched back from 2.4.12 to 2.4.8 because it is much more stable for our apps.
I am not a kernel hacker, but I make heavy use of many of my system's resources. My target systems have a least 2 Processors, 2GB of RAM, the lastet NVIDIA GPU and fast SCSI disks, and I try to render 3D scenes fast, utilizing every processor, using a lot of IPC, while asynchronously loading hundreds of megabytes per minute from disks and the network.
However, everything I program runs outside the kernel, as a user mode process, and my understanding of one of the main tasks of the kernel is to isolate (misbehaving) user processes, but it seems to me it was never so easy to take out my LINUX-boxes as it with the current 2.4.x kernel series; my app runs fine for days, but after I quit it there is a 50/50 chance the system survives. I get kernel oopses or the system freezes completely; sometimes just the X-Server hangs; when the system continous to operate, sometime the gcc freaks out with internal compiler errors, which disappear after reboot. And I had to switch from Reiserfs to ext2 because of serious filesytem corruptions, and NFS is even more troublesome than usual under LINUX
My system logs are full of kernel messages like:
Sep 25 15:02:21 dellomat kernel: Unable to handle kernel paging request at virtual address 002200dc
Sep 25 15:02:21 dellomat kernel: Unable to handle kernel NULL pointer dereference at virtual address 00000000
Oct 30 18:13:38 dellomat kernel: kernel BUG at page_alloc.c:81!
I am not absolutely sure I can blame the kernel, but I can rule out the hardware, as the problems appear on at least five different platform on a about ten systems. And the severity of these problems varies with kernel versions; the most stable we currently use is 2.4.8, and 2.4.12 was almost unusable for us.
Before reading the article, my favorite suspect for causing these problems were the NVIDIA drivers, but now everything points to some vital parts of the kernel causing these problems. I do really appreciate the great work many people are doing on the kernel, and I think I can't criticise someone for doing me a favour by working on a free OS, but I am really concerned that all this might seriously endanger the only alternative OS that runs one a broad range of state-of-the-art hardware.
I hope that there are not too much people having the same trouble, and I will get rid of these problems without becoming a kernel hacker.
While writing this I became aware of the "Linux Test Project", and I am going to download and run a bunch of tests now; I never thought this would be necessary as I do not do kernel development, but is seems to be the best thing I can do towards getting rid of the problems. (Except whining on slashdot, of course;-)
...after a series of filesystem corruption on four different Machines using different Versions of ReiserFS with many different Kernels from 2.4.2 to 2.4.12, with different SCSI disks as well as on several IDE drives, and systems ranging from a Dell Inspiron 8000 Notebook over some homegrown single PIII, dual PIII's on different Mobos to a Dell dual P4 Rambus system. For the last twenty years I have never seen something like this:
After power cuts on frozen development systems it regularly happened that files written minutes ago were completely corrupted; they were there, but just garbage in them; what you have written explains what probably happened; however, it troubled me that files written minutes ago were affected. What really upset me to throw out ReiserFS on every machine was when after a crash every File I created within the last two hours was destroyed; I never thought a Filesystem might take out many hundred files with such a precision. Even if I would not blame ReiserFS for this disaster (I Do), I consider it as completely unacceptable that all this happened without the slightest warning; no entry in the syslog, no boot message, nothing. ReiserFS pretended everything is fine. Do you have any explanation for such an behaviour, and are such effects just the downside for using a journaling fs, or is it something ReiserFS specific ? What added to my loss of confidence into this ReiserFs was that a few months ago reiserfsck did core dump when I tried to repair a file system that showed strange behaviour, which I regarded as exceptional behavior at that time.
For now I switched back to ext2 and feel pretty good to see a thorough filesystem check after a crash. I do not remember much trouble using XFS with IRIX, but I have no experience so far with any journaling fs on linux exept those mentioned above. So do You have any recommendation for a filesystem on a unstable development system, where I can not sacrifice too much performance, but need at least confidence into the integrity of my fs ? (I did not loose much data, but It easily takes a few hours to bring back a system from the backups, but an unnoticed damage to vital files can drive you crazy).
p.
Slashdot Mirror
Microsoft had more than 1000 people working on Internet Explorer alone.
The 10-20 LOC/day rate also includes a lot meetings and excessive documentation before, during and after coding, and endless debugging sessions.
Working 40 hours per week almost alone in a two-people office without phone on a familiar task with guru level programming language skills some people can churn out complex tested high quality code at a sustained rate of few hundred lines per day.
On the other hand, if you are sitting with fifty other people in one room with frequent interruptions, several six hour meetings per week, coordinating with some hundred foreign programmers in different timezones around the world trying to implementing something specified in this six-thousand page document nobody really understands, working under high pressure twelve hours every day including weekends. you will easily find your output below 10 LOC/day.
p.
Sometimes, especially in spring, when I walk around in the woods and watch brown and yellow leafs, sometimes some colors feel strange, like an itching inside my brain. Once I stared for several minutes into some square meters of ground covered with yellow leafs reflecting the direct light of the sun. I was happy and excited just by looking at this color, which in fact was not like any yellow I have seen before in my life. It seemed to be yellow, orange, brown and green at the same time. The sensation felt great, and I after a while looking at it I was sure this color could never be reproduced on an RGB display. Sometimes I also feel slightly disoriented when bathed in red or blue light from flourescent lamps, which are placed by artists in some public places here, and I like this sensation of brain-tickling so I often to walk through there just to get this strange feeling.
Regarding MP3, when I started to listen to 128kbit MP3, I was hardly able to recognize a difference. Later I became very sensitive to the artfacts of MP3-compression and almost stopped to listen to 128kBit MP3 for some types of music, especially Madonna and The Beatles sound really bad to me when compressed at 128kbit. I could not hear artifacts with 128 kBit OggVorbis Compression, but I am curious whether I will after a while. Also this effect of feeling bad while listening to an MP3 Versions seems to happen with music I heard very often ucompressed before.
The article does not give any answers, but it raises some really important questions. However, I am afraid they will never be answered, because after more than fifty years of television the consequences are still not understood very well.
p.
yellow-brown-orange
The wrong idea that sending people into space is science and research goes all back to the Apollo program, when after the first landing on the moon NASA tried to sell subsequent missions as "scientific missions".
IMHO, sending a man to the moon was the highest cultural achievement of mankind in history so far, but as a piece of art, there is no much value in repeating it, and as nobody had the balls to admit that hundred billion dollars were spent for art, it had to be science.
There is plenty of important science that happens in space, but you don't need people hanging around for that.
Manned Space Exploration is about beeing there, and to feel how it feels to be there. It is about living there. It is about building houses, planting trees and fathering children out there. And cruising around with a cool car, if you are american.
After Apollo 17 all space programs world started to decline, and there is no end in sight. The Space Shuttle program started by crippling Wernherr von Braun's original design that had a piloted, horizontal landing reuasable first stage by using a cheap throw-away fuel tank and reusable solid fuel boosters, ending up with a Space Shuttle with more expensive payload than using throw-away rockets. The buerocrats way of wasting money by budget cutting. And every news I heard about the ISS the last twenty years was about budget cuts and delays. I heard you need 2.5 people just to operate it, and there are three guys up there. SNAFU.
It is sad, and I hope I will be wrong, but within a decade we will see:
- The ISS will be abandoned a finally reenter the atmosphere
- The last Space Shuttle will go out of service
- There will be no more capabilities to send humans into orbit any more
I just hope mankind will regain manned space travel before we will have depleted our natural resources here down on earth.In this particular case you have just four criterias in claim 1, and the are pretty unspecific, so it it is a patent possibly dangerous to many people. There are two additional independent claims 4 and 7, which you can view as different additional claims that were put into the patent to widen it's scope. The rest of the patent just clarifies and specializes these independent claims.
It is the examiner's job to narrow the claims as much as possible, and the applicant usually want to have them as wide as possible. Here, definitely the applicants did a better job than the examiner.
From what I see, there is no real invention here, but that is true for most of the so called IT-Patents, and this one is not a particularly bad example, it is merely a typical patent you often have to write because the competition does it too.
p.
Regarding changing of priorities, I think that with SCHED_OTHER the priority is beeing automatically modified by the scheduler to distibute cycles in a more fair fashion.
I tried both SCHED_RR and SCHED_FIFO and changing priorities basically works, but it seemed to me that changing priorities did not have an immediate effect as required to implement priority ceiling locks.
For example, when boosting the priority of a thread to the ceiling priority, and the thread is the only one with this priority, I expect it to run without beeing preempted by anyone before the priority is lowered or the process blocks. On the other hand, when lowering the priority, I expect a higher prio thread to be executed immediately. I would also expect the order of unblocking threads is correctly adjusted when their priority was changed while suspended.
However, it seems that priority changes do not much affect the actual timeslice or the unblocking order, but I did not have the means to find out what exactly happens; using a debugger is outright impossible with fine-grained multi threaded programs.
Is it possible that some system thread needs to run inbetween to do some housekeeping ? Do you have any hints about the scheduler's inner workings ?
Thank you
p.
scheduler does not immediately respond to priority changes
thread-specific storage access is slow
There is a well known effect in multi-threaded programming called priority inversion that can cause deadlocks when a low-priority thread has acquired a resource that a high priority thread is waiting for, but a medium priority thread keeps the low priority thread from beeing executed and so the medium priority thread effectively gets more cycles than the high priority thread.
One way to overcome this problem is to use priority ceiling locks where the priority of a thread is boosted to a ceiling value when it acquires a lock. Unfortunately I found that changing the priority of a thread for a short interval does not have any effect at all with the current 2.4.x standard pthreads implementation.
The second problem I ecountered is that accessing thread-specific storage with pthread_getspecific() takes 100-200 processor cycles on 1 Ghz PIII, which makes this common approach to overcome hotspots almost as slow as locking.
Does anyone know if any of these issues are adressed by the new implementations ?
p.
I made exactly the opposite experience; after switching to XFS, we had no problems with corrupted filesystems any more, on about 30 different development machines.
I will never use ReiserFS again because this trouble I had.
p.
From the engineering point of view, you have to drop some legacy from time to time, or either your progress will slow down, or you need more horsepower to keep your speed, like the Microsoft IE development with more than 1000 people, or like the Windows XP dev team beeing 10-30 times larger than the Windows NT team.
And it is perfectly in the tradition of Apple, in contrast with the Microsoft/Intel tradition of keeping A20 gates and VGA modes more than ten years after nobody really uses them any more.
Apple would either require a new OS 9 release for every new Mac, or had to keep the current system design, unable to benefit from new processor or bus technologies.
I have been a long time Mac User and developer between '85 and '99, but Apple lost me to Windows and Linux when fucked up their new OS development (Rhapsody etc.). I would give them a try when the gcc compiler on their top-of-the-line laptops would be at least 50% as fast as on a current Dell top-of-the-line system under Linux. (It was five times slower the last time I tried.)
p.
p.
I participated in the test myself, and it was a true blind test; it was really hard to tell between the three best sounding files at 128kbit, but there were definetely three codecs better than the rest at 128 kbit. And after receiving the results, I was glad that I could trust my ears, as I correctly identified the uncompressed version at 64 kbit and put it 2nd at 128 kbit.
I am really annoyed by MP3 artifacts, and I observed that I became more sensitive to them during the years of listening MP3s. This coincides with blindly placing MP3 on the bottom in both tests.
I am really excited about the outcome and it will have two consequences for me:
1)I will get an OGG Encoder/Player asap and encode my CDs in OGG from now on.
2) I will enjoy listening to compressed music much more than to date as I know now that I wasn't able to tell the difference from the original at 128k.
Here are my personal test result, and at 128k I was already a bit tired, and I could not really find a difference between the top 3:
Ihre Bewertung für 64 kBit/s-Codecs.:
Platz 1: unkomprimiert (WAV)
Platz 2: AAC
Platz 3: Ogg Vorbis
Platz 4: MP3Pro
Platz 5: RealAudio
Platz 6: Windows Media Audio
Platz 7: MP3
Ihre Bewertung für 128 kBit/s-Codecs:
Platz 1: Windows Media Audio
Platz 2: unkomprimiert (WAV)
Platz 3: Ogg Vorbis
Platz 4: AAC
Platz 5: RealAudio
Platz 6: MP3
Platz 7: MP3Pro
p.
On SIGGRAPH last month I was told by ATI Engineers that the new Linux Drivers from ATI will not be Open Source; they will follow NVIDIA in this. I have no experiece with the current open source drivers, but six months ago there was *no* hardware T&L support a.k.a. 3D hw accelleration with these drivers.
p.
1) Try to make sure that your overtime does not come for free, or you will be doomed. Ask for compensation in free time after the project will be over, and make a written agreement about the terms, exact dates. Propose a reasonable cutoff like 2 hours per day and 8 hours on weekends. Do not agree on additional money.
2) The best time to talk about this is when start a job; talk about overtime policies explicitly, how much is excpected during normal times, during a crisis, and how much you are willing to give under these circumstances.
3)Buy, read and cite "Peopleware" and "The Deadline" to your managers; "Extreme Programming" even goes that far to have a "No Overtime" policy as a part of the process rules.
4) As as Team Leader, you have to fight for your Team and protect it from impossible demands from upper management, even risking to be fired.
5) Divide the Project into 20 to 50 tasks, and let everone on the team estimate the required time for every task; present the anonymized results to the management.
6) When there is crisis, productivity is not as importing as successfully meeting the deadline; every person on the team has to find individually the point where the most work is beeing done per time, while "Work Done" means 100% completed functionality meeting the required defect rates, not 90% ready bug ridden pieces. Depending on age, type of work, and current personal condition this optimal point lies between working 6-11 hours daily, at 5-6 day a week.
7) Typical Management often is not as mentally exhausting like development; Large meetings, travelling, reading and writing messages can feel like spare time compared to complicated design, coding or debugging sessions.
8) Try to avoid getting into thinking "them" against "me/us". Offer every support and idea to meet the business objectives.
9) You have the right on a chance to be successful. Make yourself aware and explain to your management that they are denying you this right, and take consequences if it does not change: Either quit or pick and stick to an elaborated "doomed project survival strategy".
10) The is no such thing as a free lunch. Overtime has many associated cost; quality, productivity, health, well-being at work; make sure that the bill goes the one who ordered.
"A mother needs nine months to give birth to a child, and even four mothers can't make it in five months. And contrary to the belief of some managers, even twenty mothers can't do it faster." [Kent Beck, IIRC]
p.
Java was supposed to be a small, open platform for writing clients that are delivered through the web browser.
Java was never supposed to run in a Browser or the Internet at all; see a short history here. It was started 1991 without any special purpose in mind and was then focussed 1993-1994 to run on these "Interactive Television" set top boxes. After this failed there were some plans to use it as CD-ROM/online multimedia platform. Finally two guys wrote a Browser in Java (remember HotJava?), and only in 1995 Netscape licensed it for the Browser.
IMO Java is far from dead anywhere and is here to stay, as many mediocre technologies that were present at the right time with a successfully balanced set of tradeoffs. My language of choice is C++ and programming Java is like typing with one Hand in my pocket and half of my brain asleep, but it is the best choice for a beginner's language.
C# has no justification between Java and C++, and as a Visual Basic substitute it might succeed only in the long run.
Microsoft does not have the power to kill a programming language or guarantee it's success, as the end user does not care about what programming language something is written in, and the programmer will pick and stay with the one that he feels comfortable with.
p.
"DS9: People hanging around in a shopping mall and waiting for aliens to show up. But isn't that basically what all americans are doing ?"
(from the book "All I needed to know about life I learned from Star Trek")
p.
Honestly, it is easy possible to find configurations where I can prove either view. In general, it is still a pain to get all hardware supported and configuered under linux; wlan, firewire, cameras and high-end audio devices are just a few examples I usually spend days to make them work properly.
p.
In the late eigthies, you still had to write almost every application if you required something like a workflow and wanted the system to be used by mere mortals.
Today, one person alone can set up an integrated office system with document management, customer database, accounting, reporting, E-mail access, firewalls, printing, backup, telephone integration, banking, video and audio editing, ftp and webserver, dns, dhcp, wavelan, gigabit backbone, raids, computing clusters and mobile vpn access within weeks, and several hundredthousand apps are available for free or very little money. You can easyly afford a database software or an office-suite worth a few hundred million dollars of develepment effort, and get even millions of man -months delivered on a $50 Linux Distribution, including the source code. Or if you buy a Mac or a PC today, you can actually do useful things without even installing additional software - remember, what you could do with an AppleII out of the Box ?
So today, if you want to roll your own stuff, or even spend a few days on improving your customer database access, you need many thousand customers to justify even using a real database instead of MS Outlook contacts or a simple spreadsheet.
Trying to build a new 3D-Engine, a Web-Browser, a database engine or a new GUI library is almost insane from a business point of view, so the deeper you descend into the swamps of IT-Development, the better your justification has to be for shelving out that money and taking the risk of failure.
I try to ride Moore's law and aim for something unique that does not exist so far because it was impossible or too hard to do in the past. I try to stay current by spending a lot of time hands on new technology, and I steadyly improve my and my team's knowledge and skills, but I admit: it is increasingly harder to find and exploit those niches where you have both: Fun and Profit.
OTOH, there will always be the Linus way: Build something for the sheer fun and knowledge, and the worst case is that you are happier and smarter afterwards.
And if you don't mind to listen to someone who has been around for a while and covered some distance: Never do things for the only purpose of profit. It will minimize your chances, but even if you succeed, you will not be any happier than today. And probably have fewer friends.
parabyte
Fine joke, except: There is no *dark* side of the moon; the moon's rotation locks one side to the earth, not to the sun, so the *far* side of the moon gets as much sunlight as the near side.
...while controlled by the autopilot is my favourite pet bug.
Or the other way around, 11 dead in five months would correspond to a community of about 3000 experts, which seems to be a probable size.
As you can see, it is all well within the normal death rate. Nothing strange here, just some math-handicapped newspeople making up a story out of nothing.
p.
It causes overweight not only because you don't move around, but it slows down metabolism while it makes you feel hungry and lets you eat tons of snacks while not beeing aware of it (->fat)
While your body goes into a sleep-like state, you feel very exhausted after long TV sessions; this really wrecks havoc on your immune system(->sick)
Watching TV before sleep forces your brain's nightly garbage collection to work on what you have seen on TV instead on your real life experiences, with negative effects on decision making and dealing with real life problems, up to causing serious mental diseases(->mad)
Watching TV makes the parts of you brain responsible for controlling your motion wither, which is known to negatively affect your intellectual capabilities (->stupid)
And finally the incredible amount of trash transmitted over U.S. TV can not have any positive effects on the spectator, and all this is regularily interrupted with a mixture of ads, many of them using sophisticated techniques to manipulate your mind. (I can not stand a U.S. TV program for more than a couple of minutes because among other things I am not used to these frequent breaks; it makes me so agressive that I have to switch it off)
I hope that one day the people responsible for this lunacy will have to pay for it.
p.
But the best was from the beginning that you simply insert it into your laptop and just drag all your music directly from your terabyte-server onto the disk, or play the stuff on your laptop when the ipaq ran out of power.
The real write rate I measured is between random 1124 and sequential 1260 kbytes/sec; read was beteen random 960 kbyte and sequential 1260 kybtes/sec, access time about 20 ms; It is much faster than anything I have seen with any USB device; the speed has never felt like a problem.
Other unexpected uses were out-of-the-box file transfer between an PowerBook and Wintel-Notebooks, and you can also store enough pr0n on such a disk without taking away much music capacity. You can also carry around a huge library of e-books without caring much about space.
So far I did not regret the investment, and I haven't seen anyone who wasn't impressed when you handed him the smallest one gigabyte harddisk in the world: it feels heavy and expensive in your hand, and my fingers still slightly sweat when they touch it.
p.
P.S.: It still always surprises me how small it is, and the PCMCIA-Adapter is very IBMish: it hides the connector when you remove the drive, and locks the cover to prevent unintended exposure of the contacts. It also comes with watertight two watertight small boxes with rubber polstered corners to safely stow away the PCMCIA-Adpater and the drive.
A small elite minority tries to define a "Body of Knowledge" consistent with their beliefs and slow down progress by carving it in stone and steel, and even seeking political help to limit access to the field for people who refuse to swallow their bullshit. I admit there is some useful stuff in it, and it all sounds very reasonable, but honestly: How many professionals with a master and 4 years of experience in the field (the target audience) are able to professionally apply a substantial portions of this "body" under the correct circumstances ? My guess is less than three out of hundred, and even they will ignore the majority of it because it is useless.
4 .txt) revealed a perfect example for this: A train derailed because of a bearing overheating that was indicated by an alarm 29 hours before the accident, but the alarm was disconnected because the crew was unable to find the alarm cause and assumed a defect in the monitoring system.
This self appointed elite was not able to set up a website that survives the slashdot effect, and if you ever worked with some of e.g. Rational products and assume that they used their own teaching and tools in constructing this crap, it becomes obvious that no amount of impressive sounding formal engineering methods and a big names lineup alone can produce a good piece of software.
What these methods are good for is to save your ass when the lawyers come, and to prove that you exactly built the shit the customer has ordered, and that it is not your fault that the system is too slow and too hard to use and that the operators should have sticked exactly to the procedures described on page 972 of the operations manual to avoid [two days downtime, loss of 6 months of transaction history, hitting the mountain, reversing the thrust in midair, melting the reactor core]
The attitude that comes out of this document drives me mad, and it has the potential to do so much harm to the evolution of the field that Microsoft seems to be a pleasure compared to these guys. They even try to borrow ethics and principles from medicine, a field where mystification and deception are a guiding principle since 2000 years, and to get a medical education in the roman empire you literally had to become a doctor's slave, being freed not before you were going to replace him.
When I was young I was a firm believer of almost everything my teachers and professors taught me; you may call me a naïve coward, but I was taught many times that it is a bad idea to question their doctrines for several reasons:
1) They are often correct
2) It is difficult to find out when they are wrong
3) If you find out they almost never admit they are wrong
4) They prefer to defend themselves by ridiculing, threatening and insulting
5) They have a lot of power by controlling your career path
There are exceptions, but you need a lot of luck or persistence to encounter one of them; I rarely did, so in a mixture of good faith and cognitive dissonance I spent a lot of time and effort on learning their doctrines.
But soon it became increasingly difficult to bridge the gap between many of those things I learned in academia and the first hand experience I gathered in various positions in the industry, ranging from designing hardware and kernel for embedded factory automation systems in the mid 80's to distributed virtual environments today.
During my professional career I encountered a number of programming languages, operating system, implemented ITU-, IEEE-, DIN-, ISO-, IETF, OMG and W3C standards, and went through some meters of scientific papers, produced at least a half million lines of code myself and guided the production of about five times as much.
And the more I learned from experience, the more livid I became at those teachers. What a waste of time it was, and what a large pile of bullshit I had been forced to swallow, and it took some time to get rid of all those misconceptions that were stuffed in my mind.
Don't take me wrong; I still believe that the scientific way of finding some truth is superior to almost everything else, but a large amount caution is required when dealing with scientific results, and much more caution when dealing with people who are paid for practicing and teaching science.
I am convinced that:
1) most scientific publications are obsolete (about 98% is my guess)
2) there is a tremendous abuse of scientific rituals in areas where science is hardly applicable (e.g. psychology, social sciences, medicine, business, software, history, politics, city planning, art)
3) you can find a bunch of scientist to support almost anything if you throw enough money in
4) real progress in science is rare, comes by surprise and often ruins the life of the contributors
5) most scientist are impostors
Now, coming to the engineers, who are supposed to apply science to build something useful, I am think that:
1) Engineers either think that they are the better scientists or do not know anything about science
2) most engineers are cowards
Now, when it comes to computers and process of creating software, I surmise that after gathering 25 years of experience, learning from failures and steadily improving my skills I feel qualified to raise my voice on this issue.
A simplified, unbalanced summary of my convictions regarding software is:
1) Heavyweight formal methods are a useless waste of time if want to successfully create software that satisfies you and your customer, except in extremely rare cases
2) There is no substitute for own experience, even if you can gain a lot from the experience of others
3) There is no substitute for thorough testing
4) It is helpful to have a solid mathematical knowledge, but you must not confuse math with software
5) Standardization is a necessary evil, and I commiserate those brave souls involved
6) Most standards are flawed because they are too complex to be fully understood even by their creators, not to mention mere mortals
7) Every implementation of a standard I have ever seen was flawed
8) The future belongs to light-weight, easy-to-understand iterative methods
Regarding all those cited mission-critical systems in avionics and telecommunication, most of them are of so incredible bad quality that it is a miracle that mankind survived so far; but even if the system works as designed, people often get killed because the interface between the human operator and the system does not work as expected; just browse through an arbitrarily chosen risks digest, and the first one I just picked (http://www.infowar.com/iwftp/risks/Risks-19/19_9
NASA and all other space agencies loose rockets and satellites all the time, several passenger planes were lost because of interface problems, and if you have ever watched an air traffic controller in front of his computerized radar system you will be surprised about the amount of wrong information on his screen which he has to filter out mentally.
Now these areas are where formal methods were applied for many years now, but the effort is extraordinary and the results are not convincing.
I don't need to mention the poor quality of today's popular operating systems (Linux is not better here than Windows). The most popular compilers of one of the most popular programming languages (C++) are so full of bugs that their usefulness is severely affected, but if you try to use development software that is less popular you feel like moving through a mine field: every new move you make might blow it up. It seems to be a general rule that software tends to be as poor as tolerated, no matter how bright the developers are, and only widespread use of the software improves quality.
And those products from the formal methodologist camp are not a nibble better, on the contrary: Rational Rose, for example: After so many years they still can't properly display text, zooming in the diagram views is the worst I have ever seen in any app, and it crashes more often than anything from Microsoft. (And that Rose doesn't understand C++ is no miracle; no program does)
Formal Methods are like communism: In theory, they sound good. In practice, they do not take into account the diversity of human desires and thinking, and the gap between the software engineering community and the people developing software is widening.
I would appreciate this SWEBOK if these guys were not so serious about making life harder for many people and trying to force their "body" down of throat of thousands of students.
They are marginally improving quality in some areas while hampering possible significant progress that comes from widespread use of light weight methods like Extreme Programming, which is the greatest advance in Software Engineering I have seen in my career so far.
p.
Creating Software in fact is a discipline requiring more artistic than mathematic skills; the best analogy in the "old" world is the Architect who designs houses; he also needs to be familiar with math and needs to know a lot about physical properties of different materials, but the real issue is to create a building where people like to live in,which requires to know more about people and what makes them feel comfortable, more about the history of art than the history of stone and steel.
Creating Software is about making a computer useful to people. Software is not for Computers; they feel equally comfortable with any piece of code; software is about modeling real world entities, and about creating interfaces to the human mind.
Even if you write a protocol stack or a kernel, something no normal user is exposed to, you are creating an interface that makes it possible for another programmer's mind to deal with it.
So it is not the silicon side of software, but the human mind side that makes the difference.
And now throwing in the multi-media aspects of today's most software it becomes obvious that in fifty years creating software will be considered as much an engineering discipline like making a movie today is considered as an engineering discipline because it is an opto-chemical process.
Btw: ever seen somthing funny, beautyful, lovely or cool come out IEEE ?
p.
From the artical (my Translation:)
"Possibly soon German judges will have to deal with the question whether intentional violation of the Red Book Standard by manipulating the TOC is a criminal use of 'incorrect or incomplete data'. It will be also important wether digitally reading an Audio CD into computer memory is considered as data processing."
The German Criminal Law has in the section "fraud" a the special Paragraph 263a about "computer fraud", which I translate as follows(IANAL, but I am married to one):
263a StGB Computer fraud
(1) Who damages the assets of another person with the intention of providing himself or a third person an illegal pecuniary advantage by the fact that he affects the result of a data processing procedure by incorrect design of the program, by use of incorrect or incomplete data, by unauthorized use of data or otherwise by unauthorized interference with the process is punished with imprisonment up to five years or with fine.
This paragraph is usually used against people tampering with ATMs or a company's computer systems, but I see no reason why my personal computer should not be protected by this law, and for me it is unquestionable that all these CD copy prevention systems are feeding incomplete an incorrect data to my computer, and reading a CD into the memory of my computer is definitely a data processing procedure.
As I know the courts, proving damage is often the crucial point. I think that a clear label stating "This CD is intentionally damaged so that it does not play on most computers and some CD players." would probably save the publisher, but anything short of this IMO constitutes fraud.
In this case there was no warning sign on the CD, so when I buy such a CD I do not get what I pay for. If I bought the CD just to listen to it on my MP3 Player or use a duplicate in my car stereo (my CDs do not last very long when using them in the car), then the value of the CD is zero. And even if I get a refund, it probably takes more than an hour to manage it all, which is a significant amount that easily surmounts the value of the CD: I am tricked, my assets are damaged. This is even simple fraud under 263 StGB, and even a particularly serious case under subsection 3 (repeatedly defrauding a large number of people), which is punished with imprisonement up to ten years.
I am very curious about the outcome of this case, but I would be happy if courts would stop greedy corporations trying to screw me with intentionally damaged products.
If not, I will react like many people: I will stop to buy any music at all; I will fill up a terabyte disk with enough music for the rest of my life in one afternoon, and those record companies can fuck themself.
p.
I am not a kernel hacker, but I make heavy use of many of my system's resources. My target systems have a least 2 Processors, 2GB of RAM, the lastet NVIDIA GPU and fast SCSI disks, and I try to render 3D scenes fast, utilizing every processor, using a lot of IPC, while asynchronously loading hundreds of megabytes per minute from disks and the network.
However, everything I program runs outside the kernel, as a user mode process, and my understanding of one of the main tasks of the kernel is to isolate (misbehaving) user processes, but it seems to me it was never so easy to take out my LINUX-boxes as it with the current 2.4.x kernel series; my app runs fine for days, but after I quit it there is a 50/50 chance the system survives. I get kernel oopses or the system freezes completely; sometimes just the X-Server hangs; when the system continous to operate, sometime the gcc freaks out with internal compiler errors, which disappear after reboot. And I had to switch from Reiserfs to ext2 because of serious filesytem corruptions, and NFS is even more troublesome than usual under LINUX My system logs are full of kernel messages like:
Sep 25 15:02:21 dellomat kernel: Unable to handle kernel paging request at virtual address 002200dc
Sep 25 15:02:21 dellomat kernel: Unable to handle kernel NULL pointer dereference at virtual address 00000000
Oct 30 18:13:38 dellomat kernel: kernel BUG at page_alloc.c:81!
Call Trace: [__delete_from_swap_cache+126/132] [__free_pages+27/28] [delete_from_swap_cache_nolock+106/108] [free_page_and_swap_cache+115/196]
I am not absolutely sure I can blame the kernel, but I can rule out the hardware, as the problems appear on at least five different platform on a about ten systems. And the severity of these problems varies with kernel versions; the most stable we currently use is 2.4.8, and 2.4.12 was almost unusable for us. Before reading the article, my favorite suspect for causing these problems were the NVIDIA drivers, but now everything points to some vital parts of the kernel causing these problems. I do really appreciate the great work many people are doing on the kernel, and I think I can't criticise someone for doing me a favour by working on a free OS, but I am really concerned that all this might seriously endanger the only alternative OS that runs one a broad range of state-of-the-art hardware.
I hope that there are not too much people having the same trouble, and I will get rid of these problems without becoming a kernel hacker.
While writing this I became aware of the "Linux Test Project", and I am going to download and run a bunch of tests now; I never thought this would be necessary as I do not do kernel development, but is seems to be the best thing I can do towards getting rid of the problems. (Except whining on slashdot, of course ;-)
p.
After power cuts on frozen development systems it regularly happened that files written minutes ago were completely corrupted; they were there, but just garbage in them; what you have written explains what probably happened; however, it troubled me that files written minutes ago were affected. What really upset me to throw out ReiserFS on every machine was when after a crash every File I created within the last two hours was destroyed; I never thought a Filesystem might take out many hundred files with such a precision. Even if I would not blame ReiserFS for this disaster (I Do), I consider it as completely unacceptable that all this happened without the slightest warning; no entry in the syslog, no boot message, nothing. ReiserFS pretended everything is fine. Do you have any explanation for such an behaviour, and are such effects just the downside for using a journaling fs, or is it something ReiserFS specific ? What added to my loss of confidence into this ReiserFs was that a few months ago reiserfsck did core dump when I tried to repair a file system that showed strange behaviour, which I regarded as exceptional behavior at that time.
For now I switched back to ext2 and feel pretty good to see a thorough filesystem check after a crash. I do not remember much trouble using XFS with IRIX, but I have no experience so far with any journaling fs on linux exept those mentioned above. So do You have any recommendation for a filesystem on a unstable development system, where I can not sacrifice too much performance, but need at least confidence into the integrity of my fs ? (I did not loose much data, but It easily takes a few hours to bring back a system from the backups, but an unnoticed damage to vital files can drive you crazy). p.