I've been reading this whole thread with some confusion, as I print to a Brother (1170?) via a Solaris lpd print server, and CUPS is fantastic.
I'd rather not have to use my browser to configure it, or have it opening ports without saying so explicitly during the "make install", but I've had absolutely no problem with it. It's got no Brother driver, I just print it to a raw queue, and the Solaris print server farms the job out to the printer. As Plug'n'play as plug'n'play can be.
Re:Just give me a kernel that turns off my compute
on
Linux 2.4.18 Released
·
· Score: 2
General Setup -> APM BIOS Support -> Use Real Mode APM BIOS Call To Power Off
I've been writing a utility recently which generates reports in RTF format. Unfortunately, RTF is a Microsoft format, and a document cannot be independently verified. It's basically a rehash of the internal Word format, so it's just about as messy as the internal Word format.
RTF was not developed as a language from the ground up, it's just a way of saving MS Word documents in a non-binary format.
If a language definition came along for RTF, that developers could know for sure that their application was generating valid RTF, it'd be a lot easier for everyone.
Invalid RTF can easily crash MS Word (just don't close a table, Word dies...) because it's such a hotchpotch format, even Word cannot test the file for validity before it crashes.
Whilst RTF is useful for interoperability, it's not that much better than MS Word format in some cases. An OpenDoc standard, which MS adhered to, would save everyone megabucks and megastress.
- If Microsoft takes office in a direction you don't like, you won't be tied to it.
If that happens, we can change then. Right now I don't see a good reason to.
By then it'll be too late - if you decide in 2 years time not to use MS Office, and you change to using RTF format now, you can still read all those documents you've saved over the past two years.
If you drop MS Office instantly, you've lost all your data.
Of course, promoting RedHat in Linux magazines is not necessarily preaching to the choir - like the tobacco-advertising argument (which has *some* truth, but is not the *only* truth), RedHat could turn SuSE users by advertising in a Linux magazine.
Of course, the tobacco argument is that advertising doesn't make people start using the (generic) product, just to switch from brand X to brand Y.
To make people start using the (generic) Linux, mass-market advertising would be necessary...
Re:Cox on governments adopting open source softwar
on
Alan Cox Interview
·
· Score: 1
There's no difference in my mind between saying "WTF's that.DOC document?" and "WTF's that?.SDW document?" - both (effectively) require a particular application. Requiring submissions to be in HTML, or RTF (yeah, rtf's a MS format) means that everybody can share documents... something as big as a government could make a big difference in educating a country.
Okay, so most of us don't share pre-formatted documents with our governments every day, but the news media do; If the UK Gov't rejected.DOCs from the BBC, requesting say.RTFs instead, the BBC would be sure to do an article or two about it on the radio, and probably on their prime TV news spot. That'd get the discussions going, and increase awareness that.DOC is not the only format.
This is so very true... many end-users say that "My PC has Windows 95"... "Mine has Windows 98"... and for many of them, there's no point getting a later version of Windows (note I didn't say "upgrade") due to the increased bloat in later versions.
So when they say "My PC has Win95", it has not occurred to them that it could run Win98, or ME, XP, or even Linux. It came with 95, so it has 95. End of story.
Pre-installation of Linux on end-user desktop PCs sold in major retail outlets would have a huge impact on the total number of Linux users... but that is probably a long way off..
Of course, there's nothing (AFAIK) stopping a vendor selling a PC which they'd paid the MS Tax on, but which to they'd added/replaced Linux. It'd cost them the MS Tax plus the development work in tuning it to their end-users' needs, which in the cut-throat world of PC sales is prohibitive.
I've done the exact same thing as the poster of this article - and it took ages (weeks, IIRC) to get off the list, despite being "clean" for all that time.
One item of spam had been sent through our server, I spotted the problem, fixed it, and got told that I'd been blacklisted. I then applied to be retested ("oh please Mr. Self-Appointed Cop, please say that I am good"), and was not removed from the list for a long long time. It should be automatic. Maybe test that server once a day for the next few weeks to make sure that it stays closed, if you feel such an urge. But everybody loses when the lists are not updated promptly - the admins of previously-open relays cannot send email, innocent recipients of email from the previously-open relay don't receive email they were expecting, and the maintainers of systems using the blacklists lose faith in the accuracy of the list, and stop using them (hopefully!).
I really don't know why people bother using these lists - I've not seen anyone claim here that they've benefitted significantly from doing so, and many people are harmed.
If you have a pirated copy of Office, or any software, you have no moral highground from which to accuse MS of impropriety. Pirating MS software is common, but that does not make it any less illegal.
In this case, it's not (for me, anyway) that it's non-GNU - heck, I don't even use GNOME (nor KDE, I use XFree and IceWM), it's the idea of sending so many Linux users down a road which is following MS - an un-Open corporation (and the rest, but the not-open is the important factor).
This would encourage developers to only develop for.NET, and thus reduce the number of applications for non-MS users (since many Open Source projects are Gnome-based, they'd 90% go.NET, add the number of MS-only-but-we-follow-the-MS-plan programmers, 99.99% of apps would be.NET applications)
That would leave me out in the cold, just because I want to run Linux, but not Gnome. That's not freedom.
Tokeized fucking BASIC?!!! Shit, I've fallen through a timewarp back to 1985, which was when I last thought BASIC was a good idea.
And a current desktop system should implement this WHY? So that it can run apps designed for a PocketPC, which will be obsolete in a few years' time? And in the meantime GNOME has sent all its ships down the dead-end.NET canal never to be seen again.
Is.NET a dead end?
All new MS creations are TRANSIENT, just a necessary step to get customers to the next Big Thing - Win95 got rid of 3.1, WinNT became Win95-like, so developers distribute 32bit Win32 apps, so Win2000 and WinME can carry people through WinXP to Win.NET. I'm not saying all this was planned in advance, of course, but.NET is just a step along the way - it will be implemented just well enough to force developers to write for it, but very few people are crying that Linux doesn't run Win32 apps (okay, it can, but not nicely).
Sending GNOME down a dead-end such as.NET only guarantees that GNOME will be dead as soon as.NET is dead - and MS plan for obsolesence, otherwise how will they sell their NextBigThing(TM) in 2007?
And when customers have specific demands, you have to comply. You can't just stuff a UNIX os down their throat against their will.. (that is probably one of the fastest ways to lose your customer).
Whereas letting them stay with an insecure OS is a slow, painful death... advise them about what they need, don't just give them what they think they want - or what are they paying you for?! And whatever you end up giving them, secure it! Behind a firewall, but also on-box (and don't forget that far more cracks are done internally than from external sources).
I've got to admit that I've never used Win2k or XP, but I've heard a *lot* of comments like the grandparent post - that it just doesn't crash.
Sure, there's lots of security issues which are unrelated to the base OS's stability (Outlook, for example), but calling Windows insecure because Outlook is insecure is like calling Linux insecure because of a bad userland application (and there are tons of them on freshmeat!)
The difference is only in the top-level design, that in Windows every application has access to the whole filesystem and all the hardware, though this appears to be changing with XP (but installing 3rd party sw as a user seems to break most sw not designed with this in mind)
Yes, but if Linus is dropping stuff - even from maintainers - something needs to be fixed. There's no one tree to patch against at the moment.
All that is being suggested, is that the patch penguin accepts many patches which Linus may like, and reject the rest *with a reason*. But the patch penguin doesn't burn out because he's not making the big decisions, and his decisions don't really matter at the end of the day. He's just providing a tree for people to write patches against.
True Windows emulation...
on
Lindows Reviewed
·
· Score: 5, Funny
"Not only did the installation process not give us the opportunity to add users other than root, it
didn't even explain that we *should* add users other than root, didn't tell us that the account
was root, and even tended to discourage us from entering the optional security password for
root, because, "if you lose this password it cannot be recovered." "
"trying to run Windows programs in a user account will cause problems."
Wow, even emulating Windows' very own security model...
Most commercial Unices have dealt with this issue years ago - and not just because rebuilding the kernel is impossible without the sources, I remember relinking DRS/NX kernels without the source.
Under Solaris, I can add, for example, a tape device, rdnexus (HW RAID) device, or anything else, in some cases reboot (but not always *have* to reboot), and be up and running again, with the new device up and working.
Linux is pretty brain-dead when it comes to this - and apart from scheduling processes, what else is a kernel supposed to do? And the recent VM wars have shown that the scheduling isn't that great, either.
Sure, Linux is free, but after you've done supporting devices, the next stage in growing up, is to work seamlessly with those devices, not needing the user to hack around for a few hours, and take a few reboots of new kernels until the right configuration is found.
I love Linux, but it really does need to grow up. ESR's suggestion doesn't address this at all, he's looking in the wrong place (again). Without making Linux a microkernel (like it probably should be), modular support should not be such a ballache.
Now that'd make Windows cost-effective! Maybe MS should encourage this!
Re:GPL - Contents of the "new" license
on
Borland Backs Down
·
· Score: 1
Having downloaded the 29MB file to get the license, I can confirm:
12. AUDIT. During the term of this License and for one
(1) year thereafter, upon reasonable notice and during
normal business hours, Borland or its outside auditors will
have the right to enter your premises and access your
records and computer systems to verify that you have paid
to Borland the correct amounts owed under this License
and determine whether the Products are being used in
accordance with the terms of this License. You will
provide reasonable assistance to Borland in connection
with this provision. You agree to pay the cost of the audit
if any underpayments during the period covered by the
audit amount to more than five percent (5%) of the fees
actually owed for that period.
and also:
If you have Kylix Open Edition and you would like to
distribute your Works other than under the GNU General
Public License, then you must purchase the appropriate
commercial license from Borland
So there seems to be no change in the license whatsoever
Not necessarily a NIC - though most people would buy one, the SunFire chassis does not include a network port, you've got to get a PCI NIC to go in it.
All I am saying, is that there's not necessarily a network on *every* Sun box, even though most people would obviously include one where it's not part of the base system
The main controversial items in the license were:
1) The right to visit your systems to check that licenses are being obeyed
2) That OpenKylix could only be used for GPL development
Which of the two have they "backed down" from?
Can I now use OpenKylix to develop a non-GPL application?
Slashdot Mirror
I'd rather not have to use my browser to configure it, or have it opening ports without saying so explicitly during the "make install", but I've had absolutely no problem with it. It's got no Brother driver, I just print it to a raw queue, and the Solaris print server farms the job out to the printer. As Plug'n'play as plug'n'play can be.
General Setup -> APM BIOS Support -> Use Real Mode APM BIOS Call To Power Off
Works for me (unheard of laptop, 2.4.7)
./configure && make && make install is far preferable to ./configure ; make ; make install!
RTF was not developed as a language from the ground up, it's just a way of saving MS Word documents in a non-binary format.
If a language definition came along for RTF, that developers could know for sure that their application was generating valid RTF, it'd be a lot easier for everyone.
Invalid RTF can easily crash MS Word (just don't close a table, Word dies...) because it's such a hotchpotch format, even Word cannot test the file for validity before it crashes.
Whilst RTF is useful for interoperability, it's not that much better than MS Word format in some cases. An OpenDoc standard, which MS adhered to, would save everyone megabucks and megastress.
If you drop MS Office instantly, you've lost all your data.
Of course, the tobacco argument is that advertising doesn't make people start using the (generic) product, just to switch from brand X to brand Y.
To make people start using the (generic) Linux, mass-market advertising would be necessary...
Okay, so most of us don't share pre-formatted documents with our governments every day, but the news media do; If the UK Gov't rejected .DOCs from the BBC, requesting say .RTFs instead, the BBC would be sure to do an article or two about it on the radio, and probably on their prime TV news spot. That'd get the discussions going, and increase awareness that .DOC is not the only format.
So when they say "My PC has Win95", it has not occurred to them that it could run Win98, or ME, XP, or even Linux. It came with 95, so it has 95. End of story.
Pre-installation of Linux on end-user desktop PCs sold in major retail outlets would have a huge impact on the total number of Linux users... but that is probably a long way off..
Of course, there's nothing (AFAIK) stopping a vendor selling a PC which they'd paid the MS Tax on, but which to they'd added/replaced Linux. It'd cost them the MS Tax plus the development work in tuning it to their end-users' needs, which in the cut-throat world of PC sales is prohibitive.
Of course, I could be wrong!
One item of spam had been sent through our server, I spotted the problem, fixed it, and got told that I'd been blacklisted. I then applied to be retested ("oh please Mr. Self-Appointed Cop, please say that I am good"), and was not removed from the list for a long long time. It should be automatic. Maybe test that server once a day for the next few weeks to make sure that it stays closed, if you feel such an urge. But everybody loses when the lists are not updated promptly - the admins of previously-open relays cannot send email, innocent recipients of email from the previously-open relay don't receive email they were expecting, and the maintainers of systems using the blacklists lose faith in the accuracy of the list, and stop using them (hopefully!).
I really don't know why people bother using these lists - I've not seen anyone claim here that they've benefitted significantly from doing so, and many people are harmed.
I'm not sure what I do is anyone's business.
If you're running a copy of Windows you didn't pay for, it's MS's business (literally)
If you have a pirated copy of Office, or any software, you have no moral highground from which to accuse MS of impropriety. Pirating MS software is common, but that does not make it any less illegal.
In this case, it's not (for me, anyway) that it's non-GNU - heck, I don't even use GNOME (nor KDE, I use XFree and IceWM), it's the idea of sending so many Linux users down a road which is following MS - an un-Open corporation (and the rest, but the not-open is the important factor).
.NET, and thus reduce the number of applications for non-MS users (since many Open Source projects are Gnome-based, they'd 90% go .NET, add the number of MS-only-but-we-follow-the-MS-plan programmers, 99.99% of apps would be .NET applications)
This would encourage developers to only develop for
That would leave me out in the cold, just because I want to run Linux, but not Gnome. That's not freedom.
Tokeized fucking BASIC?!!! Shit, I've fallen through a timewarp back to 1985, which was when I last thought BASIC was a good idea.
.NET a dead end?
.NET is just a step along the way - it will be implemented just well enough to force developers to write for it, but very few people are crying that Linux doesn't run Win32 apps (okay, it can, but not nicely).
.NET only guarantees that GNOME will be dead as soon as .NET is dead - and MS plan for obsolesence, otherwise how will they sell their NextBigThing(TM) in 2007?
And a current desktop system should implement this WHY? So that it can run apps designed for a PocketPC, which will be obsolete in a few years' time? And in the meantime GNOME has sent all its ships down the dead-end.NET canal never to be seen again.
Is
All new MS creations are TRANSIENT, just a necessary step to get customers to the next Big Thing - Win95 got rid of 3.1, WinNT became Win95-like, so developers distribute 32bit Win32 apps, so Win2000 and WinME can carry people through WinXP to Win.NET. I'm not saying all this was planned in advance, of course, but
Sending GNOME down a dead-end such as
Whereas letting them stay with an insecure OS is a slow, painful death... advise them about what they need, don't just give them what they think they want - or what are they paying you for?! And whatever you end up giving them, secure it! Behind a firewall, but also on-box (and don't forget that far more cracks are done internally than from external sources).
Sure, there's lots of security issues which are unrelated to the base OS's stability (Outlook, for example), but calling Windows insecure because Outlook is insecure is like calling Linux insecure because of a bad userland application (and there are tons of them on freshmeat!)
The difference is only in the top-level design, that in Windows every application has access to the whole filesystem and all the hardware, though this appears to be changing with XP (but installing 3rd party sw as a user seems to break most sw not designed with this in mind)
In the EU, most of us pay by the minute for local calls. So the 50K HTML-formatted spam costs me real money to download.
Yes, but if Linus is dropping stuff - even from maintainers - something needs to be fixed. There's no one tree to patch against at the moment.
All that is being suggested, is that the patch penguin accepts many patches which Linus may like, and reject the rest *with a reason*. But the patch penguin doesn't burn out because he's not making the big decisions, and his decisions don't really matter at the end of the day. He's just providing a tree for people to write patches against.
"trying to run Windows programs in a user account will cause problems."
Wow, even emulating Windows' very own security model...
I did email them about this, asking what the changes would be, but they've not replied...
Most commercial Unices have dealt with this issue years ago - and not just because rebuilding the kernel is impossible without the sources, I remember relinking DRS/NX kernels without the source.
Under Solaris, I can add, for example, a tape device, rdnexus (HW RAID) device, or anything else, in some cases reboot (but not always *have* to reboot), and be up and running again, with the new device up and working.
Linux is pretty brain-dead when it comes to this - and apart from scheduling processes, what else is a kernel supposed to do? And the recent VM wars have shown that the scheduling isn't that great, either.
Sure, Linux is free, but after you've done supporting devices, the next stage in growing up, is to work seamlessly with those devices, not needing the user to hack around for a few hours, and take a few reboots of new kernels until the right configuration is found.
I love Linux, but it really does need to grow up. ESR's suggestion doesn't address this at all, he's looking in the wrong place (again). Without making Linux a microkernel (like it probably should be), modular support should not be such a ballache.
Steve.
Now that'd make Windows cost-effective! Maybe MS should encourage this!
and also:
So there seems to be no change in the license whatsoever
Not necessarily a NIC - though most people would buy one, the SunFire chassis does not include a network port, you've got to get a PCI NIC to go in it.
All I am saying, is that there's not necessarily a network on *every* Sun box, even though most people would obviously include one where it's not part of the base system
The main controversial items in the license were:
1) The right to visit your systems to check that licenses are being obeyed
2) That OpenKylix could only be used for GPL development
Which of the two have they "backed down" from?
Can I now use OpenKylix to develop a non-GPL application?
As I'm sure you're aware, Solaris is free; it's the servers which are expensive. The 280's nice, look at the V880 too, and the 3800-6800 range.