An OS with a very fragmented 5%(10%?,15%? it wold change nothing) percent of the desktop users cannot 'embrace and extend' a single-owned OS with > 70 % of desktop users.
And I don't think anybody is trying this. What they are trying is a more subtle tactic (maybe not even done at conscious level) called infiltration. Porting open-source apps to the other platform. Augmenting interoperability (document filters, plugins,... ). Blending MS technologies on an Unix-like OS. Until there will be no more reason to not to use Linux or *BSD as desktop.
Unfortunately, it works both ways. There is a serious risk that OSS loose some of their 'design purity' (if you believe that) in the process. here will be less implementation of Good Things (tm) and more Reasonable Compromises
But no cry. This is in the nature of engineering.
/SAGA_MODE ON
And until everybody may have the sources of anything developed, there will be enough developers using their free time to implement the Good Thing, for the sake of Software Well Done.
/SAGA_MODE OFF
I hope.
Maybe it is only for the sake of parents, which now are less embarassed when kids ask them to build something with the Lego box they have just received as gift.
And nowadays with universities considering (or allready have) switching to Java as their teaching language, the problem will only get worse
If you want to teach how computers work, go for C (with some assembler).
But if you want to teach logic and algorithms, such as sorting, stacks, etc..., then higher level languages do a better job, because students are less distracted by syntax and hosekeeping problems.
Re:The real question: Why Ruby?
on
Why not Ruby?
·
· Score: 2
Ruby borrows features from Python and Perl and other languages I don't know (just like Perl and Python did).
However, it does a number of new things which I don't think could have been included in either Python and Perl interprters (wrt to Python: Full garbage collection, everithing is an object approach (including built-in types), and more...) They would have meant heavy architectural changes (Python FAQ section 6 lists a few design choices that 'could have been done differently, only it's too late now, and maybe it is better so'). And they would have been incompatible, breaking tthe existing code base and annoying the community.
That is why a new language. And if a community forms around it, it means that maybe it was needed.
I have already seen this. Just substitute RUBY with LINUX, and language with operating system.
No, I don't mean that the argument is old (though it is), but that this post smells of a template with filled-in blanks.
Anyway, since I'm here, let me bait.
Bosses of bosses don't care about which language or operating systems or other technical mantra you are using, as long as it works. When it doesn't, well, then _YOU_ are in deep sh*t, not them.
And technical managers worthing their wage know that in today world either you innovate or you die. And today Very High Level Languages (like Ruby, Perl and Python) gives to software development companies so many advantages than you cannot ignore them and hope to be around tomorrow.
It could be quite humorous if Linux application started showing up on Windows, running under emulation.
More and more OSS apps start appearing in windows version, also : Gimp, perl, python,...
I believe this is a winning move, and I hope this trend will increase. When windows user will daily use OSS applications, the switch to an open-source OS will be less thraumatic. In the meanwhile, the augmented user and developer base will allow for a better development of OSS applications.
They make more money from the sales of their weakest product than all of the Open Source companies combined for all their products.
Wrong measurement unit, I think.
Try this: "They made computer more useful for people with their weakest product than all of the Open Source companies combined for all their products."
Does not sound quite true, does it? One could argue that M$oft products for the Desktop _are_ more usable than OSS programs, but not so much.
By your reasoning (a bit stretched IMO), we have already a.NET-like platform : it is called autoconf+automake:-).
Kidding aside, they are what I call 'the hard way to portability', and have managed to port hundreds of software packages on several Unix platform (and now they are slowly but constantly creeping on the 'enemy' platform, too ).
Pity that most developers are lazy(including myself) and wants more easier way to write portable software, like Java or other byte-compiled languages.
What we fail to understand is your unability to understand that Europe do not exists (yet). There is still Italy, France, Germany, Spain and so on. With a hell of a lot of more diversification than, say, between Utah and California.
Just to reply to a stereotype with a stereotype : germans are proud and responsible people. They are still ashamed of what they did in the first half of last century, so they prefer not to declaim that they are proud of their country (but they are, and with some reason).
And, adding yet another stereotype, we spaghetti people instead already forgot that most of us where on the wrong side during WWW2. So we are less shy with our patriotism (though one could argue that there are at least three ofr four different Italies around here).
And French people still calls computers 'Ordinateurs' and software 'Logiciel'. Does that tell you enough about their patriotism?
I could go on, but I want to avoid alienating the whole Europe, less I will have to emigrate in Cina.
Being paranoic, they could use it to show how BSD-stile licence is more 'American' than GPL, and push to the "healty environment" dreamed by BG, where ideas are worked out (at zero cost) in the public domain, and then MS embrace and extends them and sells them for big cash.
Actually, according to the article Jefferson would probably have considered GPL too restrictive.
We won't let our energy get siphoned off into dead-end projects like Pine, djbdns, Minix and ipf.
There are plenty of failed (or at least still-standing) free software projects, too.
Minix and Pine are different beast. For an OS, you need a large mindshare to be successful. Opening the sources is one way (not the only one) to get it.
For an e-mail client, you might do good with a close set of dedicated developers and a licence open-enough to encourage others to send back patches and bugfixes.
Choosing the right licence depends on the kind of software: even RMS gruntingly accepts LGPL for libraries, and allowed part of Ogg Vorbis to be released in a BSD-like style.
And, mainly, choosing the licence is the Right of the Author(s) [and yes, I mean EULA, too]. If you don't like their terms, don't buy/use their software.
Not necessarly of CLR. But having a standard for extension modules interface followed by both Perl and Python (and Ruby?) would be nice. No need to re-implement three times the binding to the same library (think of FTK+ or MySQL or PostresQL or orbit or... ).
Unless the unix community can effectively market an alternative API for creating
server applications which is both as friendly for end-users and as easy to develop for
Uhm...
API for server application => CORBA ?
friendly for end-user => GTK+ and/or Qt ?
easy to develop for => python ?
I admit I know nothing of.NET. But from the tidbits I get from the news, I have the feeling that any of its functions as stand-alone is already matched in some open-source technology.
It's only the integration which lacks ( and it wouldn't be so difficult to bind GTK and Corba, for instance, to create a framework wich allows to run locally the GUI of remote applications).
My business is based upon various Open Source programs. The "community" earns my company and others a bad rap.
So you should be grateful to all these 15-year-olds which wrote the software you are using for free. And forgive them if they act _supposedly_ childish sometime.
Hint: I don't believe in the concept of "OSS community" much ( the reason being I see at least one hundred of them, bent to different and sometime conflicting goals). That is way I use the term "OSS world" instead. But using "community" the way you do might be considered offensive. Unless you mean it to, that is.
Seriously people, if Microsoft released an
MS-Apache webserver the majority of the posters on this board would be demanding blood.
Not a good case. For your example to work, Apache people should have called their work AWebServer, and then M$oft should have called its own product MSWebServer.
And in that case, personally, I would have given reason to M$oft. But may be a lawyer wouldn't.
I like all your points, but one does not stand. Killustrator might be a direct competing product
fpr Adobe(tm) Illustrator(tm). More and more OSS application are ported to non-unix platforms today ( Gimp, for example ) QT for Windows and Mac esists. It would not be too hard to port KDE libs and Killustrator to either Windows or Mac.
And I'd like to believe that everybody is free to derive products name from commomly used descriptive words, regardless of previous trademarks. It stands to reason. But reason and law have divorced many years ago, making happy the lawyers population. Therefore I won't bet on that one, too.
I agree on the basics: dictate a unique desktop is not the way to go. A common standard for desktops, like the one they are trying at www.opendesktop.org (with both KDE and GNOME people involved ) would be a Good Thing(tm), however.
However, I disagree on a couple of points:
I don't want any installer automatically putting stuff in my menus.
>br>
I do. And I'm very glad my distro (Debian) has found a standard way to work around the miriads of menu formats out there. Thanks to the dedication of its developers. Surely, a common menu format would be greatly appreciated by them.
a UNIX system you do not run programs by clicking pictures
Ah, but Linux is not Unix (neither is GNU, as they say:). You would never think of putting Unix on an embedded device, would you.
And command line requires standards, too : if every package would have put its executables in a different place (instead of/usr/bin and family) you had to spend half your life adding paths to your PATH variable (just like old DOS). So, why do you want people wast time rearranging menus every time they install something?
Cross-pollination is good. But standards would be better:)
But then, packages are more than file and archive formats. It's mainly adherence to a set of rules (like Debian Policy - I'm sure each distro has its own set), therefore until there is no agreemenmt on the rules (and maybe never) a common package format and download protocol would be of little help.
I tried (after all, that's what they pay me for) but it keeps segfaulting on me (the bastard!),
So i decided to go to./ to read and maybe write some M$oft bashing . Ah, I can feel better already... another five minutes (make it ten) and I can return to debug that d*mned program... I whish I never wrote it.
People say bad things about Microsoft on Slashdot, but the full truth is much worse. Microsoft is so abusive that I have never known or heard
about anyone who understood the complete scope of Microsoft abusiveness.
I'm not sure M$oft is more evil than other big business, say IBM or SUN. It's only more powerful. Therefore, what in the others is 'standard business practice', becomes 'intolerable abusiveness' when it's M$oft to do it ( but maybe this is just fair ).
Other than price and openness (which doesn't separate linux from *BSD, etc which are equivalents)
Thell this to all people agruing aboud GPL versus BSD Licence...:) [flames apart, they _have_ a different approach to open-source]
what separates linux from the older *NIXes?
Apparently, nothing big, because Linux designers decided to go for a well-known and working architecture (also because much OSS has been developed on ethereogenus *NIX platforms, look at what the configure scripts have to do ).
However:
the old Unices are all platform dependent, GNU/Linux is multi-platform
Old unices needed expansive hardware; GNU/Linux run on cheap PCs
Old unices were sold on niche markets, often to be operated by a an elite of lab techs in white coats; GNU/Linux bring unix power to everybody wants to mess with it (still an elite, but with a totally different attitude).
And so on...
What I like of GNU/Linux is that, while being essentialy Unix, improved the original model in many little ways.
Two examples:
kernel modules: IMO a good compromise between monolitic kernel and microkernels;
bash : a unix shell that features 'modern' command history and file completion (without messing with ! and ^ and such... )
Least but not last, it is somehow because of GNU/Linux if programmers now can have modern GUI toolkits like QT and GTK+.
A couple of thing which stays in the way of Network Computer (and.NET) acceptance:
PC nmarket is world-wide; Net PC can only work where broad-band access to the Internet is cheap and ubiquitous (in my country, which buys lots of M$oft products[and copy ever more], BB is around $500 per year, and only available in big towns ).
You are right: most of peole does only two or three things with their PC (I do four:). But there is an alternative for them to NC : appliances, i.e. light PC-that-arent-PC, which can be engineered do to the two-or-three things smootly and at zero-maintenance-costs.
Network Computers may be accepted in corporate and office LAN, where most of services are already centralized. But for this is not needed any.NET technology : a remote display product for WinTEL platform is more than enough (and IIRC they are already on the market).
Slashdot Mirror
An OS with a very fragmented 5%(10%?,15%? it wold change nothing) percent of the desktop users cannot 'embrace and extend' a single-owned OS with > 70 % of desktop users.
And I don't think anybody is trying this. What they are trying is a more subtle tactic (maybe not even done at conscious level) called infiltration. Porting open-source apps to the other platform. Augmenting interoperability (document filters, plugins, ... ). Blending MS technologies on an Unix-like OS. Until there will be no more reason to not to use Linux or *BSD as desktop.
Unfortunately, it works both ways. There is a serious risk that OSS loose some of their 'design purity' (if you believe that) in the process. here will be less implementation of Good Things (tm) and more Reasonable Compromises
But no cry. This is in the nature of engineering.
And until everybody may have the sources of anything developed, there will be enough developers using their free time to implement the Good Thing, for the sake of Software Well Done.
/SAGA_MODE OFF
I hope.
Maybe it is only for the sake of parents, which now are less embarassed when kids ask them to build something with the Lego box they have just received as gift.
If you want to teach how computers work, go for C (with some assembler). ..., then higher level languages do a better job, because students are less distracted by syntax and hosekeeping problems.
But if you want to teach logic and algorithms, such as sorting, stacks, etc
However, it does a number of new things which I don't think could have been included in either Python and Perl interprters (wrt to Python: Full garbage collection, everithing is an object approach (including built-in types), and more
That is why a new language. And if a community forms around it, it means that maybe it was needed.
No, I don't mean that the argument is old (though it is), but that this post smells of a template with filled-in blanks.
Anyway, since I'm here, let me bait.
Bosses of bosses don't care about which language or operating systems or other technical mantra you are using, as long as it works. When it doesn't, well, then _YOU_ are in deep sh*t, not them.
And technical managers worthing their wage know that in today world either you innovate or you die. And today Very High Level Languages (like Ruby, Perl and Python) gives to software development companies so many advantages than you cannot ignore them and hope to be around tomorrow.
More and more OSS apps start appearing in windows version, also : Gimp, perl, python, ...
I believe this is a winning move, and I hope this trend will increase. When windows user will daily use OSS applications, the switch to an open-source OS will be less thraumatic. In the meanwhile, the augmented user and developer base will allow for a better development of OSS applications.
Wrong measurement unit, I think.
Try this: "They made computer more useful for people with their weakest product than all of the Open Source companies combined for all their products."
Does not sound quite true, does it? One could argue that M$oft products for the Desktop _are_ more usable than OSS programs, but not so much.
Kidding aside, they are what I call 'the hard way to portability', and have managed to port hundreds of software packages on several Unix platform (and now they are slowly but constantly creeping on the 'enemy' platform, too ).
Pity that most developers are lazy(including myself) and wants more easier way to write portable software, like Java or other byte-compiled languages.
What we fail to understand is your unability to understand that Europe do not exists (yet). There is still Italy, France, Germany, Spain and so on. With a hell of a lot of more diversification than, say, between Utah and California.
Just to reply to a stereotype with a stereotype : germans are proud and responsible people. They are still ashamed of what they did in the first half of last century, so they prefer not to declaim that they are proud of their country (but they are, and with some reason).
And, adding yet another stereotype, we spaghetti people instead already forgot that most of us where on the wrong side during WWW2. So we are less shy with our patriotism (though one could argue that there are at least three ofr four different Italies around here).
And French people still calls computers 'Ordinateurs' and software 'Logiciel'. Does that tell you enough about their patriotism?
I could go on, but I want to avoid alienating the whole Europe, less I will have to emigrate in Cina.
Actually, according to the article Jefferson would probably have considered GPL too restrictive.
There are plenty of failed (or at least still-standing) free software projects, too.
Minix and Pine are different beast. For an OS, you need a large mindshare to be successful. Opening the sources is one way (not the only one) to get it.
For an e-mail client, you might do good with a close set of dedicated developers and a licence open-enough to encourage others to send back patches and bugfixes.
Choosing the right licence depends on the kind of software: even RMS gruntingly accepts LGPL for libraries, and allowed part of Ogg Vorbis to be released in a BSD-like style.
And, mainly, choosing the licence is the Right of the Author(s) [and yes, I mean EULA, too]. If you don't like their terms, don't buy/use their software.
Not necessarly of CLR. But having a standard for extension modules interface followed by both Perl and Python (and Ruby?) would be nice. No need to re-implement three times the binding to the same library (think of FTK+ or MySQL or PostresQL or orbit or ... ).
Uhm ...
- API for server application => CORBA ?
- friendly for end-user => GTK+ and/or Qt ?
- easy to develop for => python ?
I admit I know nothing ofWhat I'm missing ?
Anyway, in the other story there was plenty of quoting of M$oft EULA (and a direct link in the story header).Here for example.
I don't. Never eard of it before today.
So you should be grateful to all these 15-year-olds which wrote the software you are using for free. And forgive them if they act _supposedly_ childish sometime.
Hint: I don't believe in the concept of "OSS community" much ( the reason being I see at least one hundred of them, bent to different and sometime conflicting goals). That is way I use the term "OSS world" instead. But using "community" the way you do might be considered offensive. Unless you mean it to, that is.
Not a good case. For your example to work, Apache people should have called their work AWebServer, and then M$oft should have called its own product MSWebServer.
And in that case, personally, I would have given reason to M$oft. But may be a lawyer wouldn't.
And I'd like to believe that everybody is free to derive products name from commomly used descriptive words, regardless of previous trademarks. It stands to reason. But reason and law have divorced many years ago, making happy the lawyers population. Therefore I won't bet on that one, too.
However, I disagree on a couple of points:
Ah, but Linux is not Unix (neither is GNU, as they say
And command line requires standards, too : if every package would have put its executables in a different place (instead of
But then, packages are more than file and archive formats. It's mainly adherence to a set of rules (like Debian Policy - I'm sure each distro has its own set), therefore until there is no agreemenmt on the rules (and maybe never) a common package format and download protocol would be of little help.
So, I'll settle for cross-pollination.
I tried (after all, that's what they pay me for) but it keeps segfaulting on me (the bastard!),
So i decided to go to ./ to read and maybe write some M$oft bashing . Ah, I can feel better already ... another five minutes (make it ten) and I can return to debug that d*mned program ... I whish I never wrote it.
Much better if they make sure that all important user-space OSS packages can be natively and smootly compiled on their platform.
I'm not sure M$oft is more evil than other big business, say IBM or SUN. It's only more powerful.
Therefore, what in the others is 'standard business practice', becomes 'intolerable abusiveness' when it's M$oft to do it ( but maybe this is just fair ).
Thell this to all people agruing aboud GPL versus BSD Licence
what separates linux from the older *NIXes?
Apparently, nothing big, because Linux designers decided to go for a well-known and working architecture (also because much OSS has been developed on ethereogenus *NIX platforms, look at what the configure scripts have to do ).
However :
- the old Unices are all platform dependent, GNU/Linux is multi-platform
- Old unices needed expansive hardware; GNU/Linux run on cheap PCs
- Old unices were sold on niche markets, often to be operated by a an elite of lab techs in white coats; GNU/Linux bring unix power to everybody wants to mess with it (still an elite, but with a totally different attitude).
And so onWhat I like of GNU/Linux is that, while being essentialy Unix, improved the original model in many little ways. Two examples:
Least but not last, it is somehow because of GNU/Linux if programmers now can have modern GUI toolkits like QT and GTK+.
Network Computers may be accepted in corporate and office LAN, where most of services are already centralized. But for this is not needed any .NET technology : a remote display product for WinTEL platform is more than enough (and IIRC they are already on the market).