The key thing is the microaggressions and expulsion from a community are actually pretty distantly linked.
Pretty much everyone will inadvertently post microaggressions from time to time and competent moderators know this. I'm certainly far too culturally ignorant to get it right all the time. However in this area people are normally judged not for the original post but how they react when it is pointed out to them.
I simply cannot see how having a code of conduct based on treating other people in a respectful manner will result in discouraging desirable developers and companies from joining the community. Quite the reverse in fact!
All this rage at the python devs because someone forgot to buy a RHEL (or some other companies) subscription in order to keep security fixes rolling into legacy versions of python.
There *is* an eco-system to support long term business critical use of tools such as python. You just don't go and buy it from python.org .
Don't learn the code base. Work on something. Change something. Do something.
If you're worried about breaking stuff then read back the change very carefully and follow all the impacted variables (and data types) back in a source navigator. That is when you really start learning because there's some focus. Following the variables back also gives you an insight into how to write a test case for your new work.
However for me the biggest myth in programming is that you have to understand how it all works. You only need to understand a couple of really basic common idioms (approach to threading and thread-safety, coding style) before diving in. The new code doesn't have to be right... merely make the code base better than it was before.
Wikipedia defines royalties as "usage-based payments made by one party (the "licensee") and another (the "licensor") for ongoing use of an asset, sometimes an intellectual property.". While wikipedia is not authoritive that matchs my own understanding of royalties. By their very nature royalty payments are on going.
Thus Redhat is not paying royalties to FireStar (present tense) they have paid (past tense) money to them as a royalty-buy-out (meaning neither they, nor any downstream user, must pay royalties).
This does not, in any sense whatsoever, contradict the position that royalty encumbered standards are incompatible with the GNU GPL.
Devices based on V-USB (software only USB implementation for AVR microcontrollers) are probably worth a look because designs you can copy are so numerous.
This is one of the most versatile. I doesn't support the Apple IIc yet but the BBC joysticks had a similar capability so the only difference is likely to be in the adapter lead:
Perhaps you don't want to make your own circuit board. If so, I had a quick look at the retro's schematic and reckon you should be able to get the retro firmware running on an off the shelf board such as the one adafruit sell.
I've got one of these and its a great little board. I built a temperature logger using one and I got it running (hardware and software) in about four hours. That said I did spend another three building the programmer!
Finally there a gallery of lots of HID devices made using cheap AVR controllers:
I think the fundamental problem with the article authors point 1 is that automatic assumption that the only way to contribute to fixing a bug is fix the code.
Bearing in mind the much of the skill in fixing bugs is managing to get them to reproduce on your own machine.
This means the developers are often not to be able to reproduce the problem you are having. Without clear bug reports and a willingness to experiment with your environment and set up it is nearly impossible for developers to find your bug.
Therefore to help fix bugs you need to be willing to try test versions, generate back traces, prod at you computer at the developers request.
You can't simply expect to toss a 'crappy tool just crashed' over the wall and expect the developers to read you mind and that of your computer.
> Obviously its not really 2+2=5 but by the way C, > and possibly other languages, cast from a float > to an int, it rounds up
Sorry this is completely wrong. To quote the ANSI C standard:
'When a value of floating type is converted to integral type, the fractional part is discarded.'
What has happened in the code in the parent post is a float has been added to a float resulting in a floating type with a value >5. At this point the value is floored providing the value 5.
If you really want to demonstate what is happening try:
> Since anyone else who feels like modifying your > GPL'd code is allowed to sell/offer/distribute > Microsoft's XML standard too under their license, > I fail to see why this is hostile to the GPL > license.
This Microsoft patent licence requires you to include attribution in your source code and require all derivative works to maintain this attribtuion.
The GNU GPL does not allow additional restrictions (such as requiring attribution to be maintained) to be added to the licencing conditions.
Thus you could not distribute code licenced to you under the GPL that requires the MicroSoft patent licence. This would for example prevent you from writing a MS file format reader that requires the MS licence for AbiWord or OpenOffice (products for which you do not own the copyright).
Your choices would be to write a MS file format reader with reduced functionality that does not therefore require a MS patent licence or to write an entirely seperate work at is not licenced under the GNU GPL.
It would of course be possible to write in under a modified GNU GPL which explicitly permitted Microsoft to require attribution. The licence of this code would, however, be incompatible with the normal GNU GPL.
The flat properties of nickel based rechargeable batteries are brilliant for squeezing the last bit of power out of a battery. However they do have two drawbacks.
Firstly they often mess up the built-in battery meters of devices like PDAs because these meters are designed for alkaline batteries with significant voltage drop through their life (the later Palm devices M100, M105, M125 etc. have a magic rune that can be entered to change the mode of the battery meter).
Secondly in my view rechargeable batteries can be a little dangerous in bicycle lights and similar since the rear light can simply run out of power enexpectedly putting the rider at risk.
The major problem (and I think the flaw with Mosfet's argument) is that many of the users on the mailing list are also not the right people to lead the UI design.
By their nature the people on the mailing list are the people who have learnt to use whatever crackrock interface has been designed, it is quite likely that the even enjoyed learning it. Leaving UI design in the hands of such people is by definition not going to open the UI to a greater audience.
We should be careful about allowing our current users to lead change without listening to our potential users (or to the usability experts who have spent their careers studying the behaviour of our potential users).
Reading the answer to the question 2 and the related links to a statement on Caldera's 'contributions' I suspect Mr. Love is rather overstating what Caldera has provided.
For example cscope was made available on an open source licence *before* Calera bought SCO. Further being the first to do something is certainly not a contribution to the free software community unless it is, er how can I best put it,... free software?
>His fragile little ego has been shattered into a >million itty bitty peices, poor poor Richard.
If RMS was motivated by his ego I think Stallmanix/Linux would be a better choice. As it is I think it is totally fair to call the system GNU/Linux. The vision for what has become known as Linux came from the GNU project. While the GNU project adopted existing code such as X11 where is existed the GNU projects resources where focused on *plugging the gaps*. Face it linkers and C libraries and most of the other GNU tools are derided exactly because no-one feels they are interesting. Without the GNU project none of these not-generally-interesting tools could form the centre stone of each and every GNU/Linux system.
It certainly strikes me a very unfair that the person who provided the last piece receive so much glory. The last person to climb a mountain never makes the news for long. Armstong, Aldrin and Collins, who made history?
Re:Cut and dried Copyright violation
on
Abusing the GPL?
·
· Score: 1
Regardless of the legality/illegality issue (and I have read the later comment *and* the GPL) I don't see what advantage this would buy the posters company.
The poster says they intend to release the obfusicated source code under the GPL. In this case the first person to buy the product has the right to obtain the source, re-compile it and re-distribute it. The fact the source is obfusicated does not prevent anyone from re-compiling and distributing it.
Personally I would really like to see BT win this patent battle (no keep reading). I can see nothing that would bring home the message about how flawed the US patent system is that seeing the web forcibly turned into a proprietary system. At the very least it would give the European anti-patent lobby more clout at a time when the EU is (still) forming a policy on software patents.
Further there are drivers that exist for Linux that ARE NOT GPL. The owner of the copyright for these (usually the hardware company the driver works with) is likely to benefit from their code from being used on Solaris as well.
This is totally legitamate use for these tools. Not all the drivers ported with it will be GPLed at all.
Further it is interesting to note that "open-source" software companies have to be very heavily branded. Frequently the only edge they have over competitors is their brand name. When a company needs a new chip targeted with gcc most will phone Cygnus (Redhat) for one reason and only one reason, Cygnus is the 'leading brand' for open source compilers.
Slashdot Mirror
The key thing is the microaggressions and expulsion from a community are actually pretty distantly linked.
Pretty much everyone will inadvertently post microaggressions from time to time and competent moderators know this. I'm certainly far too culturally ignorant to get it right all the time. However in this area people are normally judged not for the original post but how they react when it is pointed out to them.
I simply cannot see how having a code of conduct based on treating other people in a respectful manner will result in discouraging desirable developers and companies from joining the community. Quite the reverse in fact!
Agree on the professional organizations...
All this rage at the python devs because someone forgot to buy a RHEL (or some other companies) subscription in order to keep security fixes rolling into legacy versions of python.
There *is* an eco-system to support long term business critical use of tools such as python. You just don't go and buy it from python.org .
This is right on the money.
Don't learn the code base. Work on something. Change something. Do something.
If you're worried about breaking stuff then read back the change very carefully and follow all the impacted variables (and data types) back in a source navigator. That is when you really start learning because there's some focus. Following the variables back also gives you an insight into how to write a test case for your new work.
However for me the biggest myth in programming is that you have to understand how it all works. You only need to understand a couple of really basic common idioms (approach to threading and thread-safety, coding style) before diving in. The new code doesn't have to be right... merely make the code base better than it was before.
Wikipedia defines royalties as "usage-based payments made by one party (the "licensee") and another (the "licensor") for ongoing use of an asset, sometimes an intellectual property.". While wikipedia is not authoritive that matchs my own understanding of royalties. By their very nature royalty payments are on going.
Thus Redhat is not paying royalties to FireStar (present tense) they have paid (past tense) money to them as a royalty-buy-out (meaning neither they, nor any downstream user, must pay royalties).
This does not, in any sense whatsoever, contradict the position that royalty encumbered standards are incompatible with the GNU GPL.
Devices based on V-USB (software only USB implementation for AVR microcontrollers) are probably worth a look because designs you can copy are so numerous.
This is one of the most versatile. I doesn't support the Apple IIc yet but the BBC joysticks had a similar capability so the only difference is likely to be in the adapter lead:
http://denki.world3.net/retro_v2.html
Perhaps you don't want to make your own circuit board. If so, I had a quick look at the retro's schematic and reckon you should be able to get the retro firmware running on an off the shelf board such as the one adafruit sell.
http://www.adafruit.com/index.php?main_page=product_info&cPath=16&products_id=174
I've got one of these and its a great little board. I built a temperature logger using one and I got it running (hardware and software) in about four hours. That said I did spend another three building the programmer!
Finally there a gallery of lots of HID devices made using cheap AVR controllers:
http://www.obdev.at/products/vusb/prjhid.html
Google says:
Your search - "superior ability of apt-get to deal with flamebait" - did not match any documents.
I think the fundamental problem with the article authors point 1 is that automatic assumption that the only way to contribute to fixing a bug is fix the code.
Bearing in mind the much of the skill in fixing bugs is managing to get them to reproduce on your own machine.
This means the developers are often not to be able to reproduce the problem you are having. Without clear bug reports and a willingness to experiment with your environment and set up it is nearly impossible for developers to find your bug.
Therefore to help fix bugs you need to be willing to try test versions, generate back traces, prod at you computer at the developers request.
You can't simply expect to toss a 'crappy tool just crashed' over the wall and expect the developers to read you mind and that of your computer.
> Obviously its not really 2+2=5 but by the way C,
> and possibly other languages, cast from a float
> to an int, it rounds up
Sorry this is completely wrong. To quote the ANSI C standard:
'When a value of floating type is converted to integral type, the fractional part is discarded.'
What has happened in the code in the parent post is a float has been added to a float resulting in a floating type with a value >5. At this point the value is floored providing the value 5.
If you really want to demonstate what is happening try:
printf("%d + %d = %d\n", (int) num, (int) num, (int) (num+num));
> Since anyone else who feels like modifying your
> GPL'd code is allowed to sell/offer/distribute
> Microsoft's XML standard too under their license,
> I fail to see why this is hostile to the GPL
> license.
This Microsoft patent licence requires you to include attribution in your source code and require all derivative works to maintain this attribtuion.
The GNU GPL does not allow additional restrictions (such as requiring attribution to be maintained) to be added to the licencing conditions.
Thus you could not distribute code licenced to you under the GPL that requires the MicroSoft patent licence. This would for example prevent you from writing a MS file format reader that requires the MS licence for AbiWord or OpenOffice (products for which you do not own the copyright).
Your choices would be to write a MS file format reader with reduced functionality that does not therefore require a MS patent licence or to write an entirely seperate work at is not licenced under the GNU GPL.
It would of course be possible to write in under a modified GNU GPL which explicitly permitted Microsoft to require attribution. The licence of this code would, however, be incompatible with the normal GNU GPL.
The flat properties of nickel based rechargeable batteries are brilliant for squeezing the last bit of power out of a battery. However they do have two drawbacks.
Firstly they often mess up the built-in battery meters of devices like PDAs because these meters are designed for alkaline batteries with significant voltage drop through their life (the later Palm devices M100, M105, M125 etc. have a magic rune that can be entered to change the mode of the battery meter).
Secondly in my view rechargeable batteries can be a little dangerous in bicycle lights and similar since the rear light can simply run out of power enexpectedly putting the rider at risk.
The major problem (and I think the flaw with Mosfet's argument) is that many of the users on the mailing list are also not the right people to lead the UI design.
By their nature the people on the mailing list are the people who have learnt to use whatever crackrock interface has been designed, it is quite likely that the even enjoyed learning it. Leaving UI design in the hands of such people is by definition not going to open the UI to a greater audience.
We should be careful about allowing our current users to lead change without listening to our potential users (or to the usability experts who have spent their careers studying the behaviour of our potential users).
What you need is a *penalty* for SLAPPing.
Look at most of the European countries where if the plaintiff loses they pay the other sides legal costs.
Being lumbered with the other sides costs really puts people off filing lawsuits where they hope the other side will just give up.
Reading the answer to the question 2 and the related links to a statement on Caldera's 'contributions' I suspect Mr. Love is rather overstating what Caldera has provided.
... free software?
For example cscope was made available on an open source licence *before* Calera bought SCO. Further being the first to do something is certainly not a contribution to the free software community unless it is, er how can I best put it,
>His fragile little ego has been shattered into a
>million itty bitty peices, poor poor Richard.
If RMS was motivated by his ego I think Stallmanix/Linux would be a better choice. As it is I think it is totally fair to call the system GNU/Linux. The vision for what has become known as Linux came from the GNU project. While the GNU project adopted existing code such as X11 where is existed the GNU projects resources where focused on *plugging the gaps*. Face it linkers and C libraries and most of the other GNU tools are derided exactly because no-one feels they are interesting. Without the GNU project none of these not-generally-interesting tools could form the centre stone of each and every GNU/Linux system.
It certainly strikes me a very unfair that the person who provided the last piece receive so much glory. The last person to climb a mountain never makes the news for long. Armstong, Aldrin and Collins, who made history?
Regardless of the legality/illegality issue (and I have read the later comment *and* the GPL) I don't see what advantage this would buy the posters company.
The poster says they intend to release the obfusicated source code under the GPL. In this case the first person to buy the product has the right to obtain the source, re-compile it and re-distribute it. The fact the source is obfusicated does not prevent anyone from re-compiling and distributing it.
Personally I would really like to see BT win this patent battle (no keep reading). I can see nothing that would bring home the message about how flawed the US patent system is that seeing the web forcibly turned into a proprietary system. At the very least it would give the European anti-patent lobby more clout at a time when the EU is (still) forming a policy on software patents.
Except Solaris 8!!!
The reason PR3 came before M18 is because M18 will ship with the PR3 bug fixes merged into it.
Further there are drivers that exist for Linux that ARE NOT GPL. The owner of the copyright for these (usually the hardware company the driver works with) is likely to benefit from their code from being used on Solaris as well. This is totally legitamate use for these tools. Not all the drivers ported with it will be GPLed at all.
Further it is interesting to note that "open-source" software companies have to be very heavily branded. Frequently the only edge they have over competitors is their brand name. When a company needs a new chip targeted with gcc most will phone Cygnus (Redhat) for one reason and only one reason, Cygnus is the 'leading brand' for open source compilers.