I always thought that the Republicans were involved in creating terrorist witches. I don't live in the states so maybe I don't understand, but that is how is appears from the outside.
Robert L Glass' book Facts and Falacies of software engineerings looks at the idea that many eyes makes bugs shallow. He concludes that there is in fact no evidence to support this.
Mozilla and the google bar block unrequested popups. If you are coding a website you should not be using unrequested popups. I think you need to take a good long look at what you are coding.
I'm am not picking on the KDE/GNOME/QT programmers. I am sure that they are very capable programmers. Maybe in many years from now I might be able to write code as well as them. The point I am making is that the GUI for a distro *must work out of the box*. Maybe not all the time (that is what patches are for), but the issues I described should not have happened with a mature distro.
My view is that what is missing from linux is a working GUI for each of the major distros. Let me explain by covering my Mandrake experiences.
MDK 8.1: made a concerted attempt to migrate from win2k to Linux. A number of the GUIs to operate basic stuff (software updates, modem install, network) didn't work. After much messing around with the command line I got most things working, but in the end I just decided to head back toi windows becuase I could work better and make changes faster.
MDK 9.2: After reading Good reviews of MDK 9.2, I thought that when giving my girlfriend a PC for Christmas, saving on the OS might be a good option and besides Linux was ready for the desktop right? MDK installed a detected all the hardware just fine, except when I went to connect to the network to download patches. The GUI to configure network card was broken (unrelated error messages, unable to configure), this was on standard and common hardware (Shuttle SN41G2, which is just an NForce 2 mobo). So fine, maybe a hardware issue with the NIC. Plug in another and the same problem. So I installed windows.
My thinking was that if I ran across this issue this early on, I would run across a stack more later. More importantly she would run across a stack more later after she has invested time and effort learning the system. I know there are fixes for the mobo problem, I don't care. The fact is the existance of the problem means that there will be others.
The way I see it Linux needs good driver support and it needs working GUIs *because that is what people use*. They don't use the command line or edit rc files. My suspicion is that because the GUI is less interesting, GUI coding is passed onto the less able coders, or is coded as an after thopught. Please note I am not talking about GNOME and KDE, I am talking about the people who use GNOME and KDE.
Speaking as a programmer, I hate writing GUI stuff. It is always the last thing I code and it is the least enjoyable part of coding. But the fact remains it is the only part of my project that normal users will see.
If linux wants to succeed on the desktop this stuff must work.
I don't think you are a bunch of blood lusting hill billies. I'm just very glad I live as far away from you as possible, in a country that has strict gun laws and a far lower crime rate.
The quote is: Of Mr Johanson's predicament, she added: "There are no rules to prevent you flying non-stop over Antarctica.
"He had enough fuel and reserve fuel based on the weather forecast and all the planning that we had done but what he encountered were winds that were not forecast by anybody."
While I don't disagree with your logic I think you have missed a key cause of dropping CD sales: Competition.
At the moment music sales are competing for the entertainment dollar in a way that they have never had to before today. DVDs, PC Games, Cinemas, the www etc we could go on all day.
Against that background I don't think that current CD prices are justified. This is more of a gut feeling than something that I can put your finger on. So locally CDs generall release at AUS$30. I consider $20 to be a fair price and I rarely buy a CD unless it is = $20.
I didn't. The hundreds of thousands of people I marched with in Sydney didn't think Iraq had WMD. Neither did the hundreds of thousands who marched all around the world. Most of the world thought Iraq didn't have WMD.
The point is your administration failed to do decent intelligence work on Iraq. They *wanted* to believe that Iraq had WMD and looked for intelligence that supported this view.
Before you dismiss me as a professional protestor, this was my first time I had been part of a protest. God willing this is the last time I will see such an apalling stupid action that I am driven to protest.
If you do a search for further information on this "somking gun" you will find that in fact the aircraft in question was used to train iraqi's in *ani-terrorism*.
I suggest you read the following (orignally published in the New Yorker) http://www.commondreams.org/views03/0506- 06.htm
It is worth noting that in my understanding the Australian law has not been tested in court when applied to CDs. In Australia it is also illegal tape TV shows. The point is that the laws are there but are not being applied.
I'm not saying I like the situation, but it isn't quite a bad as it sounds.
I think you'll find that the patent granted on the wheel was not in fact a patent as we understand patents. A new class of 'patents', innovation patents, was created. These patents are granted *without examination*. My understanding was that you would apply for one of these patents in preparation for appyling for a 'full' patent.
The guy who registered the wheel was rightly pointing out the ridulous nature of the new patents. http://www.ipmenu.com/archive/AUI_200110 0012.pdf
Download and install spam bayes and train it on worms the. There is no reason it can't be used on worms as well as spam. It catches 100% of the ~70 swen worm emails I get each day.
Incidentally the swen worm emails arrived only *after* posting to a newsgroup hosted by MS. Doesn't say much for the people living in the newsgroups.
You know if you try really hard you can put RAM in the wrong way. I plugged in a 128MB stick of SDRAM in the wrong way. I wondered why the ram wasn't being recognised and also why it was getting hot.
The parent poster is entirely correct on cutting out SQL from the FE.
The standard technique for avoiding SQL injection is to use prepared statements and stored procedures. For M$ platforms this means using the ADO Command object and stored procedures. Note that this is a comment on SQL injection, rather than any other security issues.
This also has a significant advantage in that it forces the code to be modularised. In other words data manipulation is moved to the database layer rather than the front end. In general I'd say that this will increase the code size will be matched by a decrease in errors.
Consider this, to measure a length of time (something mentioned in the article a number of times), you need a starting instant and an ending instant. Suppose you are measuring change in position over time, you need to know that starting position and ending position at those instants in time.
Point being you can't excape the instants in time.
I take it we are talking 3NF here? To suggest that a fully normalised table contains just a key-value a pair is nuts. What you are saying is that only one column can ever depend on a key. ?Que?
Simple example. Storing information about people.
PersonID [PK] FName LName
Oh dear. We've hit two rows already. It would also be perfectly valid to store a NULL in either FName or LName, as that would mean that you knew only this person's first name or their last name.
3rd normal form in summary says: 1. Atomicity, ie only 1 value in a column 2. Requires 1NF. All columns must be functionally dependant on the whole primary key rather than just a part of the key. 3. Requires 2NF. All columns should be dependant on the primary key.
NULL is a placeholder. It means that you don't know the value. It is perfectly valid when a table is fully normalised.
NULL is a perfectly valid value when you literally don't know the value for a column. The choice at that point is whether to use NULL or another, user (user being you) specified, value.
NULLs being returned in a resultset are valid when used in an OUTER JOIN, which is the situation you are talking about. In this case the relationship between the two tables might be 0 or *, ie can join 0 or an infinite number of rows.
Slashdot Mirror
I always thought that the Republicans were involved in creating terrorist witches. I don't live in the states so maybe I don't understand, but that is how is appears from the outside.
Australian currency is plastic. Try photocopying that.
Robert L Glass' book Facts and Falacies of software engineerings looks at the idea that many eyes makes bugs shallow. He concludes that there is in fact no evidence to support this.
Mozilla and the google bar block unrequested popups. If you are coding a website you should not be using unrequested popups. I think you need to take a good long look at what you are coding.
Download the google toolbar for IE.
Great idea! Then when they find the first vulnerability it will live for the life of the computer due to the fact that people can't patch easily.
Think about it.
I'm am not picking on the KDE/GNOME/QT programmers. I am sure that they are very capable programmers. Maybe in many years from now I might be able to write code as well as them. The point I am making is that the GUI for a distro *must work out of the box*. Maybe not all the time (that is what patches are for), but the issues I described should not have happened with a mature distro.
Can I add something to this?
My view is that what is missing from linux is a working GUI for each of the major distros. Let me explain by covering my Mandrake experiences.
MDK 8.1: made a concerted attempt to migrate from win2k to Linux. A number of the GUIs to operate basic stuff (software updates, modem install, network) didn't work. After much messing around with the command line I got most things working, but in the end I just decided to head back toi windows becuase I could work better and make changes faster.
MDK 9.2: After reading Good reviews of MDK 9.2, I thought that when giving my girlfriend a PC for Christmas, saving on the OS might be a good option and besides Linux was ready for the desktop right? MDK installed a detected all the hardware just fine, except when I went to connect to the network to download patches. The GUI to configure network card was broken (unrelated error messages, unable to configure), this was on standard and common hardware (Shuttle SN41G2, which is just an NForce 2 mobo). So fine, maybe a hardware issue with the NIC. Plug in another and the same problem. So I installed windows.
My thinking was that if I ran across this issue this early on, I would run across a stack more later. More importantly she would run across a stack more later after she has invested time and effort learning the system. I know there are fixes for the mobo problem, I don't care. The fact is the existance of the problem means that there will be others.
The way I see it Linux needs good driver support and it needs working GUIs *because that is what people use*. They don't use the command line or edit rc files. My suspicion is that because the GUI is less interesting, GUI coding is passed onto the less able coders, or is coded as an after thopught. Please note I am not talking about GNOME and KDE, I am talking about the people who use GNOME and KDE.
Speaking as a programmer, I hate writing GUI stuff. It is always the last thing I code and it is the least enjoyable part of coding. But the fact remains it is the only part of my project that normal users will see.
If linux wants to succeed on the desktop this stuff must work.
I don't think you are a bunch of blood lusting hill billies. I'm just very glad I live as far away from you as possible, in a country that has strict gun laws and a far lower crime rate.
You pretty much define the word "Curmudgeon" (or, dare I say, "Troll"?).
And you pretty much define the word idealist. You are minimising the problems and maximising the advantages.
Here's my guess: We won't see a man or woman on Mars within my lifetime, in other words in the next 60 years.
The quote is:
Of Mr Johanson's predicament, she added: "There are no rules to prevent you flying non-stop over Antarctica.
"He had enough fuel and reserve fuel based on the weather forecast and all the planning that we had done but what he encountered were winds that were not forecast by anybody."
Story here
While I don't disagree with your logic I think you have missed a key cause of dropping CD sales: Competition.
At the moment music sales are competing for the entertainment dollar in a way that they have never had to before today. DVDs, PC Games, Cinemas, the www etc we could go on all day.
Against that background I don't think that current CD prices are justified. This is more of a gut feeling than something that I can put your finger on. So locally CDs generall release at AUS$30. I consider $20 to be a fair price and I rarely buy a CD unless it is = $20.
I didn't. The hundreds of thousands of people I marched with in Sydney didn't think Iraq had WMD. Neither did the hundreds of thousands who marched all around the world. Most of the world thought Iraq didn't have WMD.
The point is your administration failed to do decent intelligence work on Iraq. They *wanted* to believe that Iraq had WMD and looked for intelligence that supported this view.
Before you dismiss me as a professional protestor, this was my first time I had been part of a protest. God willing this is the last time I will see such an apalling stupid action that I am driven to protest.
In fact I have heard of that.
- 06.htm
If you do a search for further information on this "somking gun" you will find that in fact the aircraft in question was used to train iraqi's in *ani-terrorism*.
I suggest you read the following (orignally published in the New Yorker)
http://www.commondreams.org/views03/0506
It is worth noting that in my understanding the Australian law has not been tested in court when applied to CDs. In Australia it is also illegal tape TV shows. The point is that the laws are there but are not being applied.
I'm not saying I like the situation, but it isn't quite a bad as it sounds.
I think you'll find that the patent granted on the wheel was not in fact a patent as we understand patents. A new class of 'patents', innovation patents, was created. These patents are granted *without examination*. My understanding was that you would apply for one of these patents in preparation for appyling for a 'full' patent.
0 0012.pdf
The guy who registered the wheel was rightly pointing out the ridulous nature of the new patents.
http://www.ipmenu.com/archive/AUI_20011
Download and install spam bayes and train it on worms the. There is no reason it can't be used on worms as well as spam. It catches 100% of the ~70 swen worm emails I get each day.
Incidentally the swen worm emails arrived only *after* posting to a newsgroup hosted by MS. Doesn't say much for the people living in the newsgroups.
You know if you try really hard you can put RAM in the wrong way. I plugged in a 128MB stick of SDRAM in the wrong way. I wondered why the ram wasn't being recognised and also why it was getting hot.
If the security vulnerability is independant of hardware is verified to exist on even MS approved hardware, surely this is not an issue.
No, it should be called zero life, as any C programmer knows that 1/2*2 = 0.
The parent poster is entirely correct on cutting out SQL from the FE.
The standard technique for avoiding SQL injection is to use prepared statements and stored procedures. For M$ platforms this means using the ADO Command object and stored procedures. Note that this is a comment on SQL injection, rather than any other security issues.
This also has a significant advantage in that it forces the code to be modularised. In other words data manipulation is moved to the database layer rather than the front end. In general I'd say that this will increase the code size will be matched by a decrease in errors.
There seems to be a flaw in the logic.
Consider this, to measure a length of time (something mentioned in the article a number of times), you need a starting instant and an ending instant. Suppose you are measuring change in position over time, you need to know that starting position and ending position at those instants in time.
Point being you can't excape the instants in time.
I take it we are talking 3NF here? To suggest that a fully normalised table contains just a key-value a pair is nuts. What you are saying is that only one column can ever depend on a key. ?Que?
Simple example. Storing information about people.
PersonID [PK]
FName
LName
Oh dear. We've hit two rows already. It would also be perfectly valid to store a NULL in either FName or LName, as that would mean that you knew only this person's first name or their last name.
3rd normal form in summary says:
1. Atomicity, ie only 1 value in a column
2. Requires 1NF. All columns must be functionally dependant on the whole primary key rather than just a part of the key.
3. Requires 2NF. All columns should be dependant on the primary key.
Really. COALESCE a bit to hard to follow?
What about OUTER JOINS?
What else are you going to use as a placeholder?
Which relational gurus (links please)?
Um no.
NULL is a placeholder. It means that you don't know the value. It is perfectly valid when a table is fully normalised.
NULL is a perfectly valid value when you literally don't know the value for a column. The choice at that point is whether to use NULL or another, user (user being you) specified, value.
NULLs being returned in a resultset are valid when used in an OUTER JOIN, which is the situation you are talking about. In this case the relationship between the two tables might be 0 or *, ie can join 0 or an infinite number of rows.
This is *not* a normalisation failure.