If the package depends on (links against) things that are GPL, the package must be released under a GPL compatable license.
To fix your example.
Crossover Office is commercial, it depends on
The linux kernel - this is GPL, but the GPL requiremnt doesn't apply to things that interact with the kernel using normal system calls.
X - is not GPL, its licensed under the X license , a modified BSD license.
libstdc - is LGPL, you can link commercial applications against LGPL libraries.
glibc - is LGPL.
Now had Crossover Office linked against a GPL library (ie readline) it would have been different,
As in all things on slashdot - the advice here is worth what you pay for it.
Back on topic.
Is the original poster intending to maintain two separate repositories or will all future work be done in the open source version ? This isn't that stupid a question - it all depends on product release mangagement - having two repositories with manual merging allows your product to "lock" its repo during key release phases while still allowing the community to continue working. While extra work - you'll be gald you did it the first time you have to release a product.
So if you only have one repo - include all the history, if you have two repos (open source and product) then I'd only export ther current HEAD revision to the open soruce repo.
Got to back this up, this was a great application. I'll be sorry to see it go.
For me (and I understand thatI'm different) it just worked - and more importantly it worked exactly how I worked. It was designed by people who knew what it meant to spend the entire day reading and writing email.
A single UI that was the same across Windows/OS X and Linux. Oh, and no crappy three paned windows rubbish that everyone else seems to want. I have a window system, why don't I let it manage the window placement ?
Greate IMAP support - this was the reason I first started to use it in 1999 on Linux - I wanted to move from mbox to IMAP and hte client I'd been using for the previous 6 years (mush/Zmail) didn't work with IMAP.
I only stopped using it a few months ago when my company moved to exchange and I had to do enough calendaring that I finally moved to Entourage, but Entourage sucks.
That the exact spec I want... a small server appliance I can run Panther Server on.
For $1000 I can get a iBook, an eMac or PowerMac G4. Neither quite fits what I want.
Something like the Cobalt Qube, but running OS X Server.
Something with
2 ethernet ports (3 preferred)
512Mb memory max
30Gb disk max
no need for keyboard/display to run, only needed for installation.
small form factor/Cube ideal or ~iBook size that was stackable,
limited OS X Server license (3 clients) to not compete with full OS X server retail product.
No expansion except firewire400, its supposed to be an appliance. Firewire required for backup.
CD-ROM only required, although firewire expansion could be dropped if DVD-R was included.
The more expensive they are the further they get from being an appliance, at $500 I'd have one for web-serving, one for mail server, one for firewall etc. At $800, I'd probably make do with just one.
WTF ? Itanium will run x86 code (albeit slower than IA64 code).
The Itanium market is small, its going to be years before it makes it to the desktop (if then), so why write an emulator for a small market when the chip already emulates the target in hardware....
At last someone who understands the difference between security and trust.
I was at the last CHATS workshop and both OpenBSD and TrustedBSD were present, along with representatives from the Linux community and other open source projects as well as commercial vendors.
While OpenBSD maybe secure, it is not trusted and will never meet the requirements for a highly trusted system (LSPP/the old B1). This is because Theo's customers don't want it (as I recall it).
Oh and can you remove the "three pane" layout rubbish. Messages should be displayed in a window that can be stacked as the user sees fit.
And I don't use folders so why bother with all that wasted screen space. But I do use the keyboard so it should be aimed at not needing a mouse.
And multiple UIs, say console and X (or web and X) for when I dial in from home.
And a scripting language (so I can do pick -t rpm-list | save +rpm-list interactively)
Zmail is still my killer application. I'd sell one of my two sisters for an industrial strength mailer than runs on Linux (mutt comes close but thye took out the best bit from mush---the shell-like interface)
You will never get A1, that requires a formal proof, and doing that on a existing code base is impossible. Doing anything higher than B1 is not worth the effort.
Most UNICES are C2 feature complete (except Linux). A few are sold in B1 versions (ie Trusted Irix)
C2 and B1 are doable. All that Linux is lacking for C2 is an audit trail. (and of course the line-by-line analysis...:-) But we can separate feature completeness from certification.
B1 requires MAC (as you stated), but to be useful it requires a MAC aware X server, commonly called CMW, that too is doable, it will just take longer, since you need some standard way of passing labels around the network.
While I'm here, can we plese separate TRUST from SECURITY. There is this crypto package that uses 1million bit keys, is it SECURE ? Yes. Can I TRUST it ? No.
richard--SGI Trust Team, but not speaking for them
So where does that leave third party applications that just want to use the rendering engine ?
If you are intending to produce a component that can be used by other apps (either by putting it in a library or via CORBA), does that mean that the third party app must also be GPL/MPL ?
GPL libraries may make RMS happy, but they have severe impact on the commercial uptake of those libraries...
This particular PST in question is known fully as CDEnext/Motifnext PST. It was set up in ~95 to fund the development of CDE/Motif 2.1
The members of the Motifnext PST are Sun, HP, IBM, Compaq(Digital), Novell, Fujitsu and Hitachi. SGI are a financial sponsor---and no I don't know what that really means.
ETC = Don't know...Executive Technology Council?
This much smaller group (3 members) was responsible for the the original funding of OSF and Motif 1.x. They own Motif up until the Motifnext PST agreement ends, at which point the funders (?) of Motifnext get ownership.
There I think I've got all that right (probably not), but I doubt you'll get a better answer unless we have an avid Open Group Legal readership. Just my understandings, not an SGI opinion.
f you're going to write a program for a year, you damn better put yourself in there Obviously you have never worked for a commercial company.
You spend 10 years writing code either for in-house use or a closed source product and no one knows your name. you write one stupid ICQ client and suddenly you're "famous".
With open source, you get paid in recognition, with closed source you get paid in money. Last time I checked, my landlord wouldn't accept glowing letters of approval saying how great a programmer I was...
There are a lot of programmers out there that for whatever reason are completely unknown in the open source world. Does that mean they have no pride in their work ? richard.
A couple of clarifies...The code known as OB1 is extracted straight from TRIX 6.5.x (our Trusted B1 product).
Its called "Sample Source" since its not a complete B1 product and if it was it could only be compiled into the Irix codebase. We thought it would help those interested in trusted systems if we made some of our code available to provide a "sample implementation" of something that is known to work.
richard.
PS. We were all laughing (well I was) that the first posting described my boss as a gentlemen...
Most of the requirements for switching depths are because most low-end PC graphics cards do not support multiple pixel formats at once.
If they did this (like SGIs have done for years), then the only reason for swithing would be to change resolutions...and that's less interesting.
Since the screen size and depth are encoded in the X protocol there would be real problems in changing this...
So you want to spend years of resouce across all the major X vendors becuase you don't have a decent monitor ? I'd be cheaper for me to just by you one:-)
Slashdot Mirror
What the hell are you smoking ?
If the package depends on (links against) things that are GPL, the package must be released under a GPL compatable license.
To fix your example.
Crossover Office is commercial, it depends on
The linux kernel - this is GPL, but the GPL requiremnt doesn't apply to things that interact with the kernel using normal system calls.
X - is not GPL, its licensed under the X license , a modified BSD license.
libstdc - is LGPL, you can link commercial applications against LGPL libraries.
glibc - is LGPL.
Now had Crossover Office linked against a GPL library (ie readline) it would have been different,
As in all things on slashdot - the advice here is worth what you pay for it.
Back on topic.
Is the original poster intending to maintain two separate repositories or will all future work be done in the open source version ? This isn't that stupid a question - it all depends on product release mangagement - having two repositories with manual merging allows your product to "lock" its repo during key release phases while still allowing the community to continue working. While extra work - you'll be gald you did it the first time you have to release a product.
So if you only have one repo - include all the history, if you have two repos (open source and product) then I'd only export ther current HEAD revision to the open soruce repo.
Got to back this up, this was a great application. I'll be sorry to see it go.
For me (and I understand thatI'm different) it just worked - and more importantly it worked exactly how I worked. It was designed by people who knew what it meant to spend the entire day reading and writing email.
A single UI that was the same across Windows/OS X and Linux. Oh, and no crappy three paned windows rubbish that everyone else seems to want. I have a window system, why don't I let it manage the window placement ?
Greate IMAP support - this was the reason I first started to use it in 1999 on Linux - I wanted to move from mbox to IMAP and hte client I'd been using for the previous 6 years (mush/Zmail) didn't work with IMAP.
I only stopped using it a few months ago when my company moved to exchange and I had to do enough calendaring that I finally moved to Entourage, but Entourage sucks.
richard.
Actaully I thought it was "If Gabriel doesn't get to roller-blade to the Chelsea piers, then the terrorists have won".
It was IXI and Visionware
Visionware orginally did an X server for Windows, IXI did Motif and X.Desktop.
IXI and VisionWare were the "Client" division and between them they did VisionFS, a clean room implementation of a SMB server.
Why would the government want to difuse it ?
If we drop the mass now in one big lump we can wipe out all the right coast Democrates in one easy move.
Then it will be Bush forever...
"Your tax dollars at work for a bright new future"
That the exact spec I want... a small server appliance I can run Panther Server on.
For $1000 I can get a iBook, an eMac or PowerMac G4. Neither quite fits what I want.
Something like the Cobalt Qube, but running OS X Server.
Something with
2 ethernet ports (3 preferred)
512Mb memory max
30Gb disk max
no need for keyboard/display to run, only needed for installation.
small form factor/Cube ideal or ~iBook size that was stackable,
limited OS X Server license (3 clients) to not compete with full OS X server retail product.
No expansion except firewire400, its supposed to be an appliance. Firewire required for backup.
CD-ROM only required, although firewire expansion could be dropped if DVD-R was included.
The more expensive they are the further they get from being an appliance, at $500 I'd have one for web-serving, one for mail server, one for firewall etc. At $800, I'd probably make do with just one.
But, but, its not even a real country...
richard.
>> Other people seem to be of the belief that having 2 competing font systems is ok. It's not - this is two competing interfaces.
s/font systems/desktops/
I suppose we should pick one of GNOME/KDE and kill the other then...
WTF ? Itanium will run x86 code (albeit slower than IA64 code).
The Itanium market is small, its going to be years before it makes it to the desktop (if then), so why write an emulator for a small market when the chip already emulates the target in hardware....
At last someone who understands the difference between security and trust.
I was at the last CHATS workshop and both OpenBSD and TrustedBSD were present, along with representatives from the Linux community and other open source projects as well as commercial vendors.
While OpenBSD maybe secure, it is not trusted and will never meet the requirements for a highly trusted system (LSPP/the old B1). This is because Theo's customers don't want it (as I recall it).
richard.
Oh and can you remove the "three pane" layout rubbish. Messages should be displayed in a window that can be stacked as the user sees fit. And I don't use folders so why bother with all that wasted screen space. But I do use the keyboard so it should be aimed at not needing a mouse.
And multiple UIs, say console and X (or web and X) for when I dial in from home.
And a scripting language (so I can do pick -t rpm-list | save +rpm-list interactively)
Zmail is still my killer application. I'd sell one of my two sisters for an industrial strength mailer than runs on Linux (mutt comes close but thye took out the best bit from mush---the shell-like interface)
richard.
I'm right there with you...The rack is prefect for what I want (and its yellow--which is big bonus).
--- richard.
Of course boxes are need to be evaluated, I can have an A1 secure OS, but if the prom allows me to bypass it, then the box isn't secure.
POI: NT has a C2 rating *including* networking.
You will never get A1, that requires a formal proof, and doing that on a existing code base is impossible. Doing anything higher than B1 is not worth the effort.
Most UNICES are C2 feature complete (except Linux). A few are sold in B1 versions (ie Trusted Irix)
C2 and B1 are doable. All that Linux is lacking for C2 is an audit trail. (and of course the line-by-line analysis...:-) But we can separate feature completeness from certification.
B1 requires MAC (as you stated), but to be useful it requires a MAC aware X server, commonly called CMW, that too is doable, it will just take longer, since you need some standard way of passing labels around the network.
While I'm here, can we plese separate TRUST from SECURITY. There is this crypto package that uses 1million bit keys, is it SECURE ? Yes. Can I TRUST it ? No.
richard--SGI Trust Team, but not speaking for them
If you are intending to produce a component that can be used by other apps (either by putting it in a library or via CORBA), does that mean that the third party app must also be GPL/MPL ?
GPL libraries may make RMS happy, but they have severe impact on the commercial uptake of those libraries...
Pot/Kettle ?
Who remembers "That's why I hate the French" a classic song from Not the Nine O'clock News.
richard
32 ? Whoop de do dah.
Even Sun can manage to scale to 64. Call me when it gets interesting.
\end{sarcasm}
This particular PST in question is known fully as CDEnext/Motifnext PST. It was set up in ~95 to fund the development of CDE/Motif 2.1
The members of the Motifnext PST are Sun, HP, IBM, Compaq(Digital), Novell, Fujitsu and Hitachi. SGI are a financial sponsor---and no I don't know what that really means.
ETC = Don't know...Executive Technology Council?
This much smaller group (3 members) was responsible for the the original funding of OSF and Motif 1.x. They own Motif up until the Motifnext PST agreement ends, at which point the funders (?) of Motifnext get ownership.
There I think I've got all that right (probably not), but I doubt you'll get a better answer unless we have an avid Open Group Legal readership. Just my understandings, not an SGI opinion.
Or rather it lies with the ETC up until the contract for PST expires when it reverts to being owned by PST.
f you're going to write a program for a year, you damn better put yourself in there Obviously you have never worked for a commercial company.
You spend 10 years writing code either for in-house use or a closed source product and no one knows your name. you write one stupid ICQ client and suddenly you're "famous".
With open source, you get paid in recognition, with closed source you get paid in money. Last time I checked, my landlord wouldn't accept glowing letters of approval saying how great a programmer I was...
There are a lot of programmers out there that for whatever reason are completely unknown in the open source world. Does that mean they have no pride in their work ? richard.
You cannot be serious.
If you don't have source you're gambling that I (as a developer of closed source/proprietory protocols) am smarter than every cracker out there ?
Even I won't take that risk...
You're well out of date. NT 4 got a C2 evaluation (with networking) in Dec 99.
Why wouldn't we want to release the source ?
Security through obscurity isn't....
Yes, thanks for doing that Tom.
A couple of clarifies...The code known as OB1 is extracted straight from TRIX 6.5.x (our Trusted B1 product).
Its called "Sample Source" since its not a complete B1 product and if it was it could only be compiled into the Irix codebase. We thought it would help those interested in trusted systems if we made some of our code available to provide a "sample implementation" of something that is known to work.
richard.
PS. We were all laughing (well I was) that the first posting described my boss as a gentlemen...
Most of the requirements for switching depths are because most low-end PC graphics cards do not support multiple pixel formats at once.
:-)
If they did this (like SGIs have done for years), then the only reason for swithing would be to change resolutions...and that's less interesting.
Since the screen size and depth are encoded in the X protocol there would be real problems in changing this...
So you want to spend years of resouce across all the major X vendors becuase you don't have a decent monitor ? I'd be cheaper for me to just by you one