I don't know what you guys are complaining about Gnome. My 5 year old (who doesn't yet know all his ABCs) feels at home with Gnome. Of course, I need to create a special setup for him, customized GDM to display his photo in login screen, removed password, removed the foot menu, added a few drawers with his favourite games. Now he can switch on the machine himself, login to his user id and start whichever game he wants.
I need not worry about him messing up the system since his user id has limited authorization. He often screws up when working on windows for running those windows based game cds, but mostly he likes working on Gnome.
This is the greatest advantage with Gnome (and Linux in general). You can set it up for a totally illiterate user who wants to do just a few things with the computer or give as much power to the power user if he wants.
Installing a web-based application is a normal use of the application. It is not a distribution. So if the web admin changes something in those scripts and does not distribute the changes, it is not a violation of GPL.
The end user is not using the software itself, but just the output produced by the software. Think that software as part of the webserver. If you are using a GPLed webserver, you can modify the server software in anyway you want, for use in your own web server, but you need not release the software to general public. Only when you distribute the modified software, do you need to release the source code.
This is same as the right to private use modifications that RMS was talking about when he objected to Plan 9 license.
We all know that corporate greed and overprotective governments are the main threat to the future of openness of the net. This battle needs to be fought in multiple points.
Educate Mr.Joe Common about the importance of independance of the net. Most of the problems occur because the general public are miseducated by the public.
Struggle to avoid enacting laws like DMCA and UCITA by educating the politicians about the problems in such laws.
Make corporations realise that how evil they are in a civilised manner. Most of the times corporations act the way they do becuase they don't even realise that their acts are affecting others.
Avoid corporate dependance on the wherever possible. For example, about 4000 Open Source projects are dependant on VA Linux for their infrastructure. (Hey, I am not saying that VA Linux is a greedy corporation. But most of the world's big companies started the same way).
Encourage artists to create content independant of corporations. Make them realise that net is an effective medium to distribute their art. From Lars interview yesterday you would have realise how little most aritists understand the net dynamics.
Document the ways to fight the corporate greed . Linux advocacy was one such well documented and most successfull efforts.
The particular technology he is refering to is Geographical Information Systems (GIS). GIS is not a satellite technology, but more like a cartographical technology. It is a database which stores information specific to the points in a map and can overlay those maps for comparative studies. Satellites images are often used in GIS systems for acquiring data. Some of the applications of GIS are, flood damage estimation, harvest forecast for different crops, land use/land cover mapping for forestry, etc.
If this were in India, we would have started a protest starting with a procession and hungerstrike in front of the offending organization (MPAA, court house or the parliament). This might not achieve the goal but will at least take the message to the common public. But I guess these kind of protests wouldn't be feasible in US.
I think what he is saying is, since Slashdot is a defendant in one of the DeCSS cases, and VA Linux now owns slashdot, will VA Linux step up the defence?
It is not just l10n that indian languages need. But good i10n too. Almost all the indian scripts fall under the category of complex scripts. That means, for example, if you want to display a unicode document for, say Hindi, a simple unicode font will not suffice. The unicode text needs to converted into a set of ligature glyphs based on a complex set of formating rules. As of date there is no API that can do this formating for Linux. Microsoft has the Uniscribe API for doing such tasks which will be available for Windows 2000.
While there are few efforts to bring similar tools into Linux world, one significant effort that needs support is the Pango library. ( http://people.redhat.com/otaylor/pango).
Unless there is a good framework that supports these complex requirements in Indian scripts localisation would not be a possibility.
I patent it. Now I want to make my OSS program, and I need LZW. Hmmm, looks like Unisys would like to use my algorithm. You scratch my back, I'll scratch yours. You let me have the rights to LZW, I let you have the rights to my algorithm.
If you are a company then this would work. But what about other open source projects? Can they use your patented algorithm without asking for your permission. If others can use your algorithm without permission, then Unisys can also use your algorithm without giving you the right to give permission to theirs?
If you want to give selective permission to open source projects, then where would you draw the line? Unisys also can probably release the program that uses your algorithm as a open source and still get away with it?
This is the basic idea around the GPL: You want my code, you must give me yours.
GPL does not say that. All it says is, if you are modifying my program and distributing the binary to others, those recipients are entitled to get a copy of the modified source code.
Ideally, what would happen is the patent origonators would sign the patents over to some organization (I'd suggest the FSF) and they would wield them to be the "viral" licenses forcing other patents to be made available to the OSS community.
Are you sure FSF is ok with that? Last time I heard FSF is against any kind of software patents. It is against the FSF philosophy.
The best way out is to lobby for abolishing software and algorithm patents.
What we need is a general purpose protocol in which the server helps clients to search contents in them. Instead of trying to emulate what napster does, we can start from scratch and design a secure protocol. This does not need to be specific to sharing MP3 files. This can be used for sharing any type of files including program binaries and other multimedia contents. Once this protocol stabilizes and is published as an RFC, it just becomes one more internet protocol like ftp and http. And then, we can have meta servers (like Archie of yester years). We can have servers and meta servers specific to content type, etc. Let us see who can block a standard internet protocol. [Of course, we may then have filters which block specific sites, etc.:-( ]
Perhaps you could change the default Anonymous Coward threshhold to something like 2, to avoid presenting ourselves as wackos to the casual reader/public?
I guess you intended to say -2. But that would disadvantage those ACs who really have worthy view points but do not want to reveal themselves. Changing the default sort order (for not logged in users) to Highest Scores First should do the trick.
I work in plain Enlightenment most of the time without any desktop environments. But on occasions, when happen to use Gnome, I go with IceWM. Enlightenment is a resource hog to use with Gnome.
When used in Gnome, IceWM does not display its own taskbar. It doesn't display its own menus on the root window, wharfs, pagers, etc. It simply decorates the windows and that is all about it.
After experimenting with atleast half a dozen other window managers, IceWM seems to be the most suitable WM to use with Gnome as of date. I haven't yet used IceWM by itself though. Should try it today!
More than a good alternative
on
GNU Inside?
·
· Score: 1
IMHO, "GNU Inside" slogan is more than a good alternative to GNU/Linux.
Many people are comfortable with saying "Linux" instead of "GNU/Linux" even when they understand and appreciate the contribution of FSF and GNU. Nobody can force them to use "GNU/Linux" instead. At the same time, nobody can force RMS/FSF/Debian not to use "GNU/Linux" either.
Besides, while saying GNU/Linux, it just becomes a name. Not everybody will dig thru the meaning of a name. It will be considered as yet another acronym by a non-geeky average Joe. But when you say 'something is inside' he will open the box and search for that 'something':-). When he doesn't find it he will try to understand what it is really.
Since the whole point of pushing on 'GNU' is to make people understand the philosophy of free software and appreciate its importance, I believe "GNU Inside" logo is a much better option than just saying GNU/Linux. And, "GNU Inside" will reach a wider audience than Linux as many have pointed out earlier.
However if the argument is just to get the credit due to FSF or RMS, then let me get out of the way!
I guess it doesn't need NIS if all the clients are Unix workstations. But when you are running DOS based PC-NFS clients there is no other way of authentication except NIS.
Though I haven't tried nfsd service on Linux, I used it on a network with multiple SCO Unix servers and pc-nfs clients (early versions of pc-nfs with 386/lesser machines on DOS). Though the performance was not anything great, it was usable. NFS is relatively easy to configure and use, but it needs NIS database for user authentication. Setting up NIS database is real pain in the neck. Is there any improvements in NIS configuration recently. Are there good tools for NIS configuration under Linux?
Probably, there is a glitch in the software that accidentally counted one vote for Gore. It is designed to count everything for Gates :-).
I don't know what you guys are complaining about Gnome. My 5 year old (who doesn't yet know all his ABCs) feels at home with Gnome. Of course, I need to create a special setup for him, customized GDM to display his photo in login screen, removed password, removed the foot menu, added a few drawers with his favourite games. Now he can switch on the machine himself, login to his user id and start whichever game he wants.
I need not worry about him messing up the system since his user id has limited authorization. He often screws up when working on windows for running those windows based game cds, but mostly he likes working on Gnome.
This is the greatest advantage with Gnome (and Linux in general). You can set it up for a totally illiterate user who wants to do just a few things with the computer or give as much power to the power user if he wants.
openofffice.org domain is owned by Caldera.
So soon we can expect another rumour that Sun is buying Caldera?
Installing a web-based application is a normal use of the application. It is not a distribution. So if the web admin changes something in those scripts and does not distribute the changes, it is not a violation of GPL.
The end user is not using the software itself, but just the output produced by the software. Think that software as part of the webserver. If you are using a GPLed webserver, you can modify the server software in anyway you want, for use in your own web server, but you need not release the software to general public. Only when you distribute the modified software, do you need to release the source code.
This is same as the right to private use modifications that RMS was talking about when he objected to Plan 9 license.
A new organization - MINC - is being setup to make domain names multilingual. If you speak a language other than English, please check it out.
-Sivaraj.
The particular technology he is refering to is Geographical Information Systems (GIS). GIS is not a satellite technology, but more like a cartographical technology. It is a database which stores information specific to the points in a map and can overlay those maps for comparative studies. Satellites images are often used in GIS systems for acquiring data. Some of the applications of GIS are, flood damage estimation, harvest forecast for different crops, land use/land cover mapping for forestry, etc.
If this were in India, we would have started a protest starting with a procession and hungerstrike in front of the offending organization (MPAA, court house or the parliament). This might not achieve the goal but will at least take the message to the common public. But I guess these kind of protests wouldn't be feasible in US.
I think what he is saying is, since Slashdot is a defendant in one of the DeCSS cases, and VA Linux now owns slashdot, will VA Linux step up the defence?
It is not just l10n that indian languages need. But good i10n too. Almost all the indian scripts fall under the category of complex scripts. That means, for example, if you want to display a unicode document for, say Hindi, a simple unicode font will not suffice. The unicode text needs to converted into a set of ligature glyphs based on a complex set of formating rules. As of date there is no API that can do this formating for Linux. Microsoft has the Uniscribe API for doing such tasks which will be available for Windows 2000.
While there are few efforts to bring similar tools into Linux world, one significant effort that needs support is the Pango library.
(
http://people.redhat.com/otaylor/pango).
Unless there is a good framework that supports these complex requirements in Indian scripts localisation would not be a possibility.
-Siva.
I patent it. Now I want to make my OSS program, and I need LZW. Hmmm, looks like Unisys would like to use my algorithm. You scratch my back, I'll scratch yours. You let me have the rights to LZW, I let you have the rights to my algorithm.
If you are a company then this would work. But what about other open source projects? Can they use your patented algorithm without asking for your permission. If others can use your algorithm without permission, then Unisys can also use your algorithm without giving you the right to give permission to theirs?
If you want to give selective permission to open source projects, then where would you draw the line? Unisys also can probably release the program that uses your algorithm as a open source and still get away with it?
This is the basic idea around the GPL: You want my code, you must give me yours.
GPL does not say that. All it says is, if you are modifying my program and distributing the binary to others, those recipients are entitled to get a copy of the modified source code.
Ideally, what would happen is the patent origonators would sign the patents over to some organization (I'd suggest the FSF) and they would wield them to be the "viral" licenses forcing other patents to be made available to the OSS community.
Are you sure FSF is ok with that? Last time I heard FSF is against any kind of software patents. It is against the FSF philosophy.
The best way out is to lobby for abolishing software and algorithm patents.
-Siva.
What we need is a general purpose protocol in which the server helps clients to search contents in them. Instead of trying to emulate what napster does, we can start from scratch and design a secure protocol. This does not need to be specific to sharing MP3 files. This can be used for sharing any type of files including program binaries and other multimedia contents. Once this protocol stabilizes and is published as an RFC, it just becomes one more internet protocol like ftp and http. And then, we can have meta servers (like Archie of yester years). We can have servers and meta servers specific to content type, etc. :-( ]
Let us see who can block a standard internet protocol. [Of course, we may then have filters which block specific sites, etc.
-Siva
Perhaps you could change the default Anonymous Coward threshhold to something like 2, to avoid presenting ourselves as wackos to the casual reader/public?
I guess you intended to say -2. But that would disadvantage those ACs who really have worthy view points but do not want to reveal themselves. Changing the default sort order (for not logged in users) to Highest Scores First should do the trick.
-Siva.
I work in plain Enlightenment most of the time without any desktop environments. But on occasions, when happen to use Gnome, I go with IceWM. Enlightenment is a resource hog to use with Gnome.
When used in Gnome, IceWM does not display its own taskbar. It doesn't display its own menus on the root window, wharfs, pagers, etc. It simply decorates the windows and that is all about it.
After experimenting with atleast half a dozen other window managers, IceWM seems to be the most suitable WM to use with Gnome as of date. I haven't yet used IceWM by itself though. Should try it today!
-Sivaraj.
If you really want pure ASCII,
do this on vi.
:1,$s/\^H./g
-Siva
If you really want pure ASCII,
do this on vi.
:1,$s/\^H.//g
-Siva
IMHO, "GNU Inside" slogan is more than a good alternative to GNU/Linux.
:-). When he doesn't find it he will try to understand what it is really.
Many people are comfortable with saying "Linux" instead of "GNU/Linux" even when they understand and appreciate the contribution of FSF and GNU. Nobody can force them to use "GNU/Linux" instead. At the same time, nobody can force RMS/FSF/Debian not to use "GNU/Linux" either.
Besides, while saying GNU/Linux, it just becomes a name. Not everybody will dig thru the meaning of a name. It will be considered as yet another acronym by a non-geeky average Joe. But when you say 'something is inside' he will open the box and search for that 'something'
Since the whole point of pushing on 'GNU' is to make people understand the philosophy of free software and appreciate its importance, I believe "GNU Inside" logo is a much better option than just saying GNU/Linux. And, "GNU Inside" will reach a wider audience than Linux as many have pointed out earlier.
However if the argument is just to get the credit due to FSF or RMS, then let me get out of the way!
-Sivaraj.
I guess it doesn't need NIS if all the clients are Unix workstations. But when you are running DOS based PC-NFS clients there is no other way of authentication except NIS.
Though I haven't tried nfsd service on Linux, I used it on a network with multiple SCO Unix servers and pc-nfs clients (early versions of pc-nfs with 386/lesser machines on DOS). Though the performance was not anything great, it was usable. NFS is relatively easy to configure and use, but it needs NIS database for user authentication. Setting up NIS database is real pain in the neck. Is there any improvements in NIS configuration recently. Are there good tools for NIS configuration under Linux?
-Siva.