I know that nobody is going to read this a week later, but anyway...
The main reason for patches in FreeBSD ports is to remove Linux specifics, and to install things into the conventional FreeBSD locations (/usr/local etc.). Two ways of looking at this: 1) if the original author of the software had done things to be more portable, this wouldn't be needed. And 2) if you are doing it for Linux, most of it shouldn't be needed since it was developed for Linux in the first place. Have a look at the patches of most modern FreeBSD ports - there just won't be any. There will just be the port system wrappers: a generic Makefile (defining dependencies, probably offering a pretty interface to build options, and often causing the use of gmake), a pkg-descr, and a pkg-plist file. Sure there are lots that don't fit this mold, but they will generally be older apps or ones that are very Linux specific. To pick a (semi-)random example that I built the other day: try gaim-devel (gaim 2.0.0b6).
Firstly, this is not an "Ask Slashdot". Secondly, I'm sure that question was purely rhetorical - Mr. Beckerman isn't really interested in Slashdot's collective wisdom!!
This can't create a new monopoly (as it currently exists) since you can save your documents to your own machine in ODF, MS, or PDF format. The monopoly is in the lack of interoperability of file formats - not in applications. That's the whole point of the ODF standard - to allow different applications to operate on a file. Google Apps goes even one further since there are even more formats available. Now, if Google prevented you from saving your documents then that would be a different matter.
The ammendments are here. I haven't read them all yet but they look fairly good. In a few places they have substituted "either for trade or to an extent that will affect prejudicially the financial interests of the performer in the performance" with "for trade". This means that non-commercial copying is no longer an offence. Also, private copying of a recording that you already own a copy of is not an offence.
I agree with what you say up to a certain point. But for anything complicated, there are always things that require some non-portable extension. The most common one I find is interfacing with third party software. On Windows, these often require the use of COM (or ActiveX, or whatever the flavour of the month is).
I run a one-man software development company in Australia. I've used Quicken Small Business Edition for the past 10 years or so. I don't like it but I've yet to find anything to replace it. Each year Quicken sends me a flyer with a discount for upgrading to version X+1. Each year, I say I'll try to find something else. In then end I go back to Quicken (either with the upgrade or just keeping my current version).
I've tried GnuCash - with version 2.0 approaching, it looks pretty good. But it's not nearly as slick as Quicken. Other alternatives are attractive for various reasons but have other problems.
My needs are very simple! I'd like multi-user access (Quicken doesn't offer this either - GnuCash may to a certain extent but it's not 100%), suitable reports for the Australian Tax Office (I know I can customise my own), fast to open and save its data files (GnuCash is very slow with 20 years of data - the multi-user requirement probably covers this because it uses Postgres), and it's go to be quick to enter invoices, payments etc. (GnuCash is horrible!). I'd prefer cross-platform and/or FOSS. If I'm going to move, it will be to something that has an open format for its data files.
if (getuid() == 0 || geteuid() != 0) do something that only root can do
What is the intention of this test? Allow something to be done if the real user id is zero (i.e. it is root logged in) or if the effective user id is not zero (i.e. if the program is running as setuid to somebody other than root). That doesn't make a lot of sense to me.
Although the ComputerWorld article mentions open source too, Peter Quinn's main battle is (was?) over open standards - a very different topic. Microsoft likes to confuse them so they can say that Massachusetts is excluding them. That's not true - they are free to implement the open standard (Open Document) in their closed source.
Strange... I've encoded all my music using Ogg Vorbis and I successfully play it on my proprietary IRiver player and through my proprietary DVR software (ShowShifter using free Directshow filters).
There's a big difference between providing a service that has a lower risk of mail being accidentally lost, and deliberately losing them unless you pay the "protection money". The latter is definitely extortion. It's got nothing to do with stopping UBE, it's purely a profit centre.
I don't believe the point of the wizard is to save keystrokes. It's because the average developer wouldn't have a clue how to do those bits of the code. They just generate it and try not to touch it. I've asked very experienced Windows developers what a particular chunk does and they just look at me blankly.
To me, any part of my code base that I don't understand is a huge risk. Sooner or later I will have to change it for some reason. I really don't want to have to do it by trial and error.
As you suggest, the biggest problem is finding good developers.
Unfortunately, that's exactly what most Visual C++ programmers mean by code generation. The "wizard" generates a huge chunk of boilerplate code that the developer never understands. He would prefer not to even look at it, but occasionally has to make some minor tweak.
When I use code generation, I generally write the generator too. As you say, there are two choices: you either generate the code once and then maintain it completely manually (i.e. it's just the scaffolding), or you maintain the generator and the input to the generator but never touch the output of the generator.
Tools like Borland C++ Builder and Rational Rose try to give you the best of both worlds with the "round trip" code generation/reverse engineering. While they do have some advantages, in my opinion they aren't overly successful at it.
That reminds me of a Yes Prime Minister sketch. The PM writes a response to a suggestion on a piece of paper "Balls!" (as in "Bullshit!"). Sir Humphrey suggests that it's a bit strong, so he changes it to "Round objects!". The reply comes back "Who is Round, and to what does he object?"
You must have very long seconds where you live! It comes out to roughly one every 6 minutes in my book (or every 1.4 minutes if he only works an 8 hour day and five days a week).
True, but you might be making the opposite assumption. Maybe the business side is much better off with Linux and it's just this one app that's the problem. The original description doesn't give enough info to be sure. Hopefully, they made that decision correctly when they went with Linux in the first place.
As much as I would like to see an accurate and bulletproof definition that prevents software-only patents from being granted, I'm not very hopeful that this will happen. One thing that I believe can dramatically reduce the effect of software patents is a definition that says that it is not possible for pure software to infringe a patent. In fact, no data stored on computer media (source code, binary code, data files, documents etc.) should be allowed to be an infringement. This is very easy to define and will prevent most of the patent misuse. If anybody is writing to MPs etc., please include this as part of your letter.
Then you haven't been looking... e.g. Australian Technology Partnerships OZtell are the two that I use. There are lots of others including Telstra and Optus, although they are mostly targetted towards big companies.
Slashdot Mirror
I know that nobody is going to read this a week later, but anyway...
The main reason for patches in FreeBSD ports is to remove Linux specifics, and to install things into the conventional FreeBSD locations (/usr/local etc.). Two ways of looking at this: 1) if the original author of the software had done things to be more portable, this wouldn't be needed. And 2) if you are doing it for Linux, most of it shouldn't be needed since it was developed for Linux in the first place. Have a look at the patches of most modern FreeBSD ports - there just won't be any. There will just be the port system wrappers: a generic Makefile (defining dependencies, probably offering a pretty interface to build options, and often causing the use of gmake), a pkg-descr, and a pkg-plist file. Sure there are lots that don't fit this mold, but they will generally be older apps or ones that are very Linux specific. To pick a (semi-)random example that I built the other day: try gaim-devel (gaim 2.0.0b6).
Hey, take that one step further and it's not a bad idea - moderation of laws. If a new law gets modded to -1, it gets taken off the books. :-)
Good luck with your case(s).
Firstly, this is not an "Ask Slashdot". Secondly, I'm sure that question was purely rhetorical - Mr. Beckerman isn't really interested in Slashdot's collective wisdom!!
This can't create a new monopoly (as it currently exists) since you can save your documents to your own machine in ODF, MS, or PDF format. The monopoly is in the lack of interoperability of file formats - not in applications. That's the whole point of the ODF standard - to allow different applications to operate on a file. Google Apps goes even one further since there are even more formats available. Now, if Google prevented you from saving your documents then that would be a different matter.
The ammendments are here. I haven't read them all yet but they look fairly good. In a few places they have substituted "either for trade or to an extent that will affect prejudicially the financial interests of the performer in the performance" with "for trade". This means that non-commercial copying is no longer an offence. Also, private copying of a recording that you already own a copy of is not an offence.
I agree with what you say up to a certain point. But for anything complicated, there are always things that require some non-portable extension. The most common one I find is interfacing with third party software. On Windows, these often require the use of COM (or ActiveX, or whatever the flavour of the month is).
I run a one-man software development company in Australia. I've used Quicken Small Business Edition for the past 10 years or so. I don't like it but I've yet to find anything to replace it. Each year Quicken sends me a flyer with a discount for upgrading to version X+1. Each year, I say I'll try to find something else. In then end I go back to Quicken (either with the upgrade or just keeping my current version).
I've tried GnuCash - with version 2.0 approaching, it looks pretty good. But it's not nearly as slick as Quicken. Other alternatives are attractive for various reasons but have other problems.
My needs are very simple! I'd like multi-user access (Quicken doesn't offer this either - GnuCash may to a certain extent but it's not 100%), suitable reports for the Australian Tax Office (I know I can customise my own), fast to open and save its data files (GnuCash is very slow with 20 years of data - the multi-user requirement probably covers this because it uses Postgres), and it's go to be quick to enter invoices, payments etc. (GnuCash is horrible!). I'd prefer cross-platform and/or FOSS. If I'm going to move, it will be to something that has an open format for its data files.
Does anybody have any clues?
Your wish is my command... VoIP wiki.
Is OpenSource a brand name?
Although the ComputerWorld article mentions open source too, Peter Quinn's main battle is (was?) over open standards - a very different topic. Microsoft likes to confuse them so they can say that Massachusetts is excluding them. That's not true - they are free to implement the open standard (Open Document) in their closed source.
Please don't confuse the two.
Congrats on being the first person on Slashdot to use the word "looser" in the correct context!
Commas, capital letters, and apostrophes in the wrong places are fucking annoying - Grammar Nazi's slogan
Considering DNA's nationality, I'd suggest that your spelling of sanitiser is wrong too.
Strange... I've encoded all my music using Ogg Vorbis and I successfully play it on my proprietary IRiver player and through my proprietary DVR software (ShowShifter using free Directshow filters).
There's a big difference between providing a service that has a lower risk of mail being accidentally lost, and deliberately losing them unless you pay the "protection money". The latter is definitely extortion. It's got nothing to do with stopping UBE, it's purely a profit centre.
I don't believe the point of the wizard is to save keystrokes. It's because the average developer wouldn't have a clue how to do those bits of the code. They just generate it and try not to touch it. I've asked very experienced Windows developers what a particular chunk does and they just look at me blankly.
To me, any part of my code base that I don't understand is a huge risk. Sooner or later I will have to change it for some reason. I really don't want to have to do it by trial and error.
As you suggest, the biggest problem is finding good developers.
Unfortunately, that's exactly what most Visual C++ programmers mean by code generation. The "wizard" generates a huge chunk of boilerplate code that the developer never understands. He would prefer not to even look at it, but occasionally has to make some minor tweak.
When I use code generation, I generally write the generator too. As you say, there are two choices: you either generate the code once and then maintain it completely manually (i.e. it's just the scaffolding), or you maintain the generator and the input to the generator but never touch the output of the generator.
Tools like Borland C++ Builder and Rational Rose try to give you the best of both worlds with the "round trip" code generation/reverse engineering. While they do have some advantages, in my opinion they aren't overly successful at it.
That reminds me of a Yes Prime Minister sketch. The PM writes a response to a suggestion on a piece of paper "Balls!" (as in "Bullshit!"). Sir Humphrey suggests that it's a bit strong, so he changes it to "Round objects!". The reply comes back "Who is Round, and to what does he object?"
One of the effects of his theft-of-thoughts is the
Don't you know anything.. it's piracy-of-thoughts.. hang on, no, no it's fair use of thoughts. Oops, sorry, wrong rant.
--
Graham
Yep, I think you're right. Cold as buggery down here!
Just pile everything on the table!
You must have very long seconds where you live! It comes out to roughly one every 6 minutes in my book (or every 1.4 minutes if he only works an 8 hour day and five days a week).
Graham
True, but you might be making the opposite assumption. Maybe the business side is much better off with Linux and it's just this one app that's the problem. The original description doesn't give enough info to be sure. Hopefully, they made that decision correctly when they went with Linux in the first place.
Graham
As much as I would like to see an accurate and bulletproof definition that prevents software-only patents from being granted, I'm not very hopeful that this will happen. One thing that I believe can dramatically reduce the effect of software patents is a definition that says that it is not possible for pure software to infringe a patent. In fact, no data stored on computer media (source code, binary code, data files, documents etc.) should be allowed to be an infringement. This is very easy to define and will prevent most of the patent misuse. If anybody is writing to MPs etc., please include this as part of your letter.
Graham
Then you haven't been looking... e.g.
Australian Technology Partnerships
OZtell
are the two that I use. There are lots of others including Telstra and Optus, although they are mostly targetted towards big companies.
Graham