But does Open Source mean that there CAN'T be a design spec or any formal methods? Just because the current leading lights don't use them, doesn't mean it CAN'T be done.
There's plenty of closed source stuff that doesn't have any formal methods - does that mean that it can't be done? No. Maybe it's harder to do with open source, because the development cycle tends to be more relaxed, and it's harder to impose methodologies on those working on it, but that doesn't make it impossible.
I can understand why it's done (although I don't agree with doing it!). Hence the question about is there a better way (and anyone who suggests using the year of release will be shot:-)
I would assume that most of the techy Linux users are above simple number comparisons, so presumably the reasoning behind playing "catch up" (and I'm not aiming this at Slackware any more than any other distribution), is for the corporate users - does it really make that much difference at the moment given that (I assume) Linux in corporates is driven by technical people more than marketing people?
And I'm only on RedHat 6.2 - I'd better upgrade to Slackware, as it's obviously a newer and better release!
Seriously though, why are the Linux distro's playing the (MS approved) numbers game? Do we really need this sort of numbering system (i.e. people jumping numbers to "keep up"). Is there a better way that actually provides a useful comparison (other than the kernel number, which in itself isn't that useful).
Does any site have a list of Distro releases, kernel versions, and major software versions (sendmail, gcc, etc)?
Well, if you don't care about his opinion, why did you read the article and comment on it? Presumably you care enough to disagree with it - in which case you are making his opininion important.
Personally I don't follow what RMS says with any particular fervour, but it is interesting to see his viewpoint, particularly on licensing matters, where he has strong views, and does set certain standards that others choose to follow.
I'm not so sure about them being doomed because of companies doing their own port. If Loki establish themselves as "the company with the know-how" then there will still be work - if you look at games that come out on multiple platforms, you will usually see that the different versions are written by different companies anyway.
someone should register microsofts.com if it hasn't been done already. Let's face it, if having an extra s is ok, how about microssoft.com and various other alternatives. Might lend a bit of weight to the argument.
"But your honour, Microsoft claim the domain microssoft.com infringes their trade mark although it is spelt differently. However they also claim that digitaldiva.com and digitaldivas.com are sufficiently different. Please let us know which case is valid."
Somewhere I think I still have a couple of copies of Windows 1.01 (or possibly 1.03) on 360k 5.25" floppies. IIRC max 4 windows, none of which could overlap.
The figure I always heard quoted was 1 help desk staff for every 25-50 users. I've never yet managed to find a company that actually has a ratio anything like this though.
From personal experience I would say it's more common to have 4-8 first line help desk staff and 2-4 second line for 500-1000 users. So at best that's 1:42 and at worst 1:167. Of course it depends on what sort of company and help desk you are talking about. These figures are from a helpdesk supporting a single site for a multinational corporation.
Surely code monkeys should be writing code, not running the show? I would prefer a great programmer spent his working time improving the code, rather than running a business. I don't really see how Open Source alters the business model that much - companies sell services rather than systems - well, that's not exactly a world shattering event, theres been plenty of service industry work for years now outside the IT world (even within it!).
All I can re-iterate is that back in the old BBS days (1987 or so onwards), most of the good programmers I knew considered themselves hackers. Until the media started using it to refer to computer crime, that WAS the meaning of the word hacker in the UK. The word cracker WASN'T used for anyone. The word phreak was used to describe those who abused the phone system, but no-one within the programming community had yet coined a phrase for people who insisted on trying to break into computers.
This debate has been going on for at least 10 years - in one debate (on FidoNet IIRC), someone suggested using the terms "stoats" and "weasels" instead. Unfortunately I could never remember which was supposed to be which.
As to 'within the scene' and 'outside the scene' what scene are you talking about. Certainly I have not met any programmer-type-hacker who has been happy with the media's use of the word hacker, and most of the people I have spoken to consider both of your examples to be 'crackers'.
Can they actually (legally) prevent him - I can understand to a certain extent the restrictions on his using technology, but to actually prevent him from talking about it to others? Is this supposed to stop him from passing on techniques to others - if so, wouldn't it be easier to require him not to incite others to follow his example? I'm not from the US, but doesn't this conflict with his constitutional rights to free speech?
No, copyright is an idea intended to protect the individual. That is why you automatically have copyright on any original works you produce unless you assign it to someone else. So, if you as an individual paint a picture, write a book, some computer software, etc. You have control over who can or cannot make copies, and under what circumstances. This is a very basic, and very powerful individual right. Patents as they currently stand benefit corporations, but not copyright.
The attitudes that copyright is somehow wrong tends to be stated by people who have never put effort into creating something. Even if you don't expect to make a profit, it is not unreasonable to want control over how your creation is distributed.
IBM were sued over their alleged monopoly in the mainframe market, not PCs.
Yes, but as part of the anti-trust IBM were not allowed to cross-subsidise any development. So OS/2 had to be priced at a point that the sales of the OS would recoup the cost of development, rather than pricing it cheaply to get market share. Also, I believe that one of the reasons that IBM used outside components for the PC (rather than developing it all in house), was they didn't want to be seen to be trying to establish an IBM monopoly on personal computing.
Somewhere I still have a Jupiter Ace... it's probably being kept company by my Sinclair QL, ZX81 and Philips CP/M machine. Shhh, I've just uncovered the fabled computer graveyard.
I'm not sure I fully understand this - so here I am, a software developer with one major commercial project that I want to port to Linux. So I get hold of copies of the major distributions, make sure my software will compile under them, and release it. I now have several versions for different platforms and (possibly) different dists on the same platform. How has this fragmented anything? If I want to release versions for commercial unix, I have to do the same thing (AIX, Solaris, SCO would all require specific versions). If I want to release for NT 4 on MIPS, i386 and Alpha I have to have different versions. How does this prevent me release closed source? Of course no-one may buy it because it's not righteous and open source, but then maybe I have a market niche and a "must-have" product. (I wish...).
There is a major difference - in Microsoft's case if you don't accept the license you have a very attractive drinks mat, unless you can convince someone to give you a refund (which appears to be either unlikely or impossible). In the case of the GPL you are entitled to use the software and re-distribute it, but you can't use sections of the source code without abiding by the license.
Yes, the GPL is intended to be viral in concept. If you don't like it, don't use it, and don't use other GPL'd code. You have a choice just like you have a choice not to accept anyone else's licensing conditions.
All this is totally irrelevant to whether or not the GPL should be considered proprietary though. You obviously consider the GPL a bad license - that's fine, despite what some people might tell you, that is your right! Personally I'm fairly GPL neutral (it achieves what it sets out to, whether you agree with those aims or not). It is not the only license, it is not the final word. However, I don't think it can be classes as proprietary.
Must be a strange form of proprietary software that allows anyone to copy the software and give it away, or to alter it for personal use. Surely the fact that it has a license doesn't make it proprietary? Or can only totally unlicensed (non-licensed?) software be free?
Why do you think this is any more impossible than have a standard graphics API that works across any number of video cards (i.e. DirectX, Direct3D etc.). As a matter of necessity games have to work with all manner of video cards providing a different variety or hardware facilities. Given that in the majority of platforms the HARDWARE supported will be the same, this shouldn't be much harder than providing the facilities for a single platform.
So let me get this straight. Microsoft wrote OS/2. Microsoft wrote Windows. IBM sued Microsoft but because Windows and OS/2 are seperate products, IBM lost.
Excuse me - if Microsoft wrote them both, it wouldn't matter if they were similar or not - you ARE allowed to infinge your OWN copyright!
Actually Microsoft and IBM collaborated on OS/2. Microsoft decided to write OS/2-lite and called it Windows. OS/2 didn't get the market share for several reasons - it was perceived as being for IBM's PS/2 range only (it wasn't) - it had (for the time) high memory requirements - it was targetted at the 286 and was outdated when the 386 was released. OS/2 since release 2.0 has been a decent stable OS that is still used by banks throughout the world. It had an object-orientated GUI before Windows9x. It has far lower hardware and memory requirements than NT (try a 486/66 with 32mb of RAM running 8 business applications simultaneously on NT).
If you see a copy of Warp on discount at your local store, buy it - even if it's just to see an alternative view of what NT might have been.
I think there are probably two main reasons why this isn't done (rather than can't be done): firstly as device drivers are responsible for the majority of I/O in a system, and unless you are number crunching, the majority of your system time is spent processing I/O, there is a need for device drivers to be as efficient as possible - this tends to rule out abstarction levels which would have an efficiency cost. Secondly, any standard API would be subject to "extension" when new facilities became available, and you would either have to wait for a standards body to incorporate them (SLOW), or deal with different vendors extending API's and the incompatabilities that would create.
Slashdot Mirror
But does Open Source mean that there CAN'T be a design spec or any formal methods? Just because the current leading lights don't use them, doesn't mean it CAN'T be done.
There's plenty of closed source stuff that doesn't have any formal methods - does that mean that it can't be done? No. Maybe it's harder to do with open source, because the development cycle tends to be more relaxed, and it's harder to impose methodologies on those working on it, but that doesn't make it impossible.
I can understand why it's done (although I don't agree with doing it!). Hence the question about is there a better way (and anyone who suggests using the year of release will be shot :-)
I would assume that most of the techy Linux users are above simple number comparisons, so presumably the reasoning behind playing "catch up" (and I'm not aiming this at Slackware any more than any other distribution), is for the corporate users - does it really make that much difference at the moment given that (I assume) Linux in corporates is driven by technical people more than marketing people?
And I'm only on RedHat 6.2 - I'd better upgrade to Slackware, as it's obviously a newer and better release!
Seriously though, why are the Linux distro's playing the (MS approved) numbers game? Do we really need this sort of numbering system (i.e. people jumping numbers to "keep up"). Is there a better way that actually provides a useful comparison (other than the kernel number, which in itself isn't that useful).
Does any site have a list of Distro releases, kernel versions, and major software versions (sendmail, gcc, etc)?
Well, if you don't care about his opinion, why did you read the article and comment on it? Presumably you care enough to disagree with it - in which case you are making his opininion important.
Personally I don't follow what RMS says with any particular fervour, but it is interesting to see his viewpoint, particularly on licensing matters, where he has strong views, and does set certain standards that others choose to follow.
I take it you haven't read the article, and probably have no intention of doing so. Go read, then post.
I'm not so sure about them being doomed because of companies doing their own port. If Loki establish themselves as "the company with the know-how" then there will still be work - if you look at games that come out on multiple platforms, you will usually see that the different versions are written by different companies anyway.
someone should register microsofts.com if it hasn't been done already. Let's face it, if having an extra s is ok, how about microssoft.com and various other alternatives. Might lend a bit of weight to the argument.
"But your honour, Microsoft claim the domain microssoft.com infringes their trade mark although it is spelt differently. However they also claim that digitaldiva.com and digitaldivas.com are sufficiently different. Please let us know which case is valid."
Somewhere I think I still have a couple of copies of Windows 1.01 (or possibly 1.03) on 360k 5.25" floppies. IIRC max 4 windows, none of which could overlap.
No, I M Weasel
Seems to be a lack of editorial control! Neither link even vaguely looks like it would lead to any specific files.
So, are stories vetted properly before they are submitted or rejected, or is it more a case of who the submitter is?
The figure I always heard quoted was 1 help desk staff for every 25-50 users. I've never yet managed to find a company that actually has a ratio anything like this though.
From personal experience I would say it's more common to have 4-8 first line help desk staff and 2-4 second line for 500-1000 users. So at best that's 1:42 and at worst 1:167. Of course it depends on what sort of company and help desk you are talking about. These figures are from a helpdesk supporting a single site for a multinational corporation.
Surely code monkeys should be writing code, not running the show? I would prefer a great programmer spent his working time improving the code, rather than running a business. I don't really see how Open Source alters the business model that much - companies sell services rather than systems - well, that's not exactly a world shattering event, theres been plenty of service industry work for years now outside the IT world (even within it!).
All I can re-iterate is that back in the old BBS days (1987 or so onwards), most of the good programmers I knew considered themselves hackers. Until the media started using it to refer to computer crime, that WAS the meaning of the word hacker in the UK. The word cracker WASN'T used for anyone. The word phreak was used to describe those who abused the phone system, but no-one within the programming community had yet coined a phrase for people who insisted on trying to break into computers.
This debate has been going on for at least 10 years - in one debate (on FidoNet IIRC), someone suggested using the terms "stoats" and "weasels" instead. Unfortunately I could never remember which was supposed to be which.
As to 'within the scene' and 'outside the scene' what scene are you talking about. Certainly I have not met any programmer-type-hacker who has been happy with the media's use of the word hacker, and most of the people I have spoken to consider both of your examples to be 'crackers'.
Can they actually (legally) prevent him - I can understand to a certain extent the restrictions on his using technology, but to actually prevent him from talking about it to others? Is this supposed to stop him from passing on techniques to others - if so, wouldn't it be easier to require him not to incite others to follow his example? I'm not from the US, but doesn't this conflict with his constitutional rights to free speech?
No, copyright is an idea intended to protect the individual. That is why you automatically have copyright on any original works you produce unless you assign it to someone else. So, if you as an individual paint a picture, write a book, some computer software, etc. You have control over who can or cannot make copies, and under what circumstances. This is a very basic, and very powerful individual right. Patents as they currently stand benefit corporations, but not copyright.
The attitudes that copyright is somehow wrong tends to be stated by people who have never put effort into creating something. Even if you don't expect to make a profit, it is not unreasonable to want control over how your creation is distributed.
IBM were sued over their alleged monopoly in the mainframe market, not PCs.
Yes, but as part of the anti-trust IBM were not allowed to cross-subsidise any development. So OS/2 had to be priced at a point that the sales of the OS would recoup the cost of development, rather than pricing it cheaply to get market share. Also, I believe that one of the reasons that IBM used outside components for the PC (rather than developing it all in house), was they didn't want to be seen to be trying to establish an IBM monopoly on personal computing.
Somewhere I still have a Jupiter Ace... it's probably being kept company by my Sinclair QL, ZX81 and Philips CP/M machine. Shhh, I've just uncovered the fabled computer graveyard.
I'm not sure I fully understand this - so here I am, a software developer with one major commercial project that I want to port to Linux. So I get hold of copies of the major distributions, make sure my software will compile under them, and release it. I now have several versions for different platforms and (possibly) different dists on the same platform. How has this fragmented anything? If I want to release versions for commercial unix, I have to do the same thing (AIX, Solaris, SCO would all require specific versions). If I want to release for NT 4 on MIPS, i386 and Alpha I have to have different versions. How does this prevent me release closed source? Of course no-one may buy it because it's not righteous and open source, but then maybe I have a market niche and a "must-have" product. (I wish...).
There is a major difference - in Microsoft's case if you don't accept the license you have a very attractive drinks mat, unless you can convince someone to give you a refund (which appears to be either unlikely or impossible). In the case of the GPL you are entitled to use the software and re-distribute it, but you can't use sections of the source code without abiding by the license.
Yes, the GPL is intended to be viral in concept. If you don't like it, don't use it, and don't use other GPL'd code. You have a choice just like you have a choice not to accept anyone else's licensing conditions.
All this is totally irrelevant to whether or not the GPL should be considered proprietary though. You obviously consider the GPL a bad license - that's fine, despite what some people might tell you, that is your right! Personally I'm fairly GPL neutral (it achieves what it sets out to, whether you agree with those aims or not). It is not the only license, it is not the final word. However, I don't think it can be classes as proprietary.
Must be a strange form of proprietary software that allows anyone to copy the software and give it away, or to alter it for personal use. Surely the fact that it has a license doesn't make it proprietary? Or can only totally unlicensed (non-licensed?) software be free?
Why do you think this is any more impossible than have a standard graphics API that works across any number of video cards (i.e. DirectX, Direct3D etc.). As a matter of necessity games have to work with all manner of video cards providing a different variety or hardware facilities. Given that in the majority of platforms the HARDWARE supported will be the same, this shouldn't be much harder than providing the facilities for a single platform.
So let me get this straight. Microsoft wrote OS/2. Microsoft wrote Windows. IBM sued Microsoft but because Windows and OS/2 are seperate products, IBM lost.
Excuse me - if Microsoft wrote them both, it wouldn't matter if they were similar or not - you ARE allowed to infinge your OWN copyright!
Actually Microsoft and IBM collaborated on OS/2. Microsoft decided to write OS/2-lite and called it Windows. OS/2 didn't get the market share for several reasons - it was perceived as being for IBM's PS/2 range only (it wasn't) - it had (for the time) high memory requirements - it was targetted at the 286 and was outdated when the 386 was released. OS/2 since release 2.0 has been a decent stable OS that is still used by banks throughout the world. It had an object-orientated GUI before Windows9x. It has far lower hardware and memory requirements than NT (try a 486/66 with 32mb of RAM running 8 business applications simultaneously on NT).
If you see a copy of Warp on discount at your local store, buy it - even if it's just to see an alternative view of what NT might have been.
I think there are probably two main reasons why this isn't done (rather than can't be done): firstly as device drivers are responsible for the majority of I/O in a system, and unless you are number crunching, the majority of your system time is spent processing I/O, there is a need for device drivers to be as efficient as possible - this tends to rule out abstarction levels which would have an efficiency cost. Secondly, any standard API would be subject to "extension" when new facilities became available, and you would either have to wait for a standards body to incorporate them (SLOW), or deal with different vendors extending API's and the incompatabilities that would create.