No, I didn't take your comments in too negative a light.
There are some activities where a centralized approach makes sense. Napster was technically a better system than Kazaa.
However, in this case as in Napster, any centralization will be the focus of an attack. A blacklist system should be designed to make an effective attack extremely difficult and expensive. I can't think of a better way to accomplish this, although it brings to mind the dreaded "p2p" word.
SQL Server will never run on any version of UNIX. AFAIK, there aren't even (MS-supported) SQL Server client libraries for non-windows platforms. I realize that FreeTDS is available, but such a library would never be used in a highly critical sector.
If you have to integrate multiple platforms, you cannot use SQL server. Closest similar product is Sybase ASE, but Microsoft broke Sybase compatibility on purpose.
I think if people understood how irrationally obstinate SQL Server's platform dependence was, they would look elsewhere. I hope that this attitude holds them below 10% penetration - it certainly has up to now. They are a bit player.
In any case, there is a cheap, new version of DB2 out for $500/copy.
Centralization of the blacklist is bad. Therefore, the lists should be p2p.
Each blacklist should be signed by the maintainer's private key. The public keys should be kept in several well-known locations.
An application, running on a mailserver, should have options to:
Download blacklists from specified upstream sources, preferably by rsync protocol, although even gzip would be an improvement over what we've had.
Apply some or all of the blacklists to inbound messages.
Offer the blacklists for further download.
Automatically announce new blacklists, the recall of canceled blacklists, or newer/faster/replacement upstream blacklist servers.
The blacklist application should work with all major MTAs, including sendmail and exchange. It should be platform-neutral, and we should do what is necessary to get MS to package it on the CD.
I can easily see web content filtering going the same way eventually.
I understand that the ESA is testing a new orbiter. Why is there no broad collaboration by NASA in this effort?
Could it be that Boeing and other major contractors behind the shuttle continually flex their political muscle to keep the shuttle program alive because of the large maintenance contracts?
NASA should find a way to shove the contractors in the closet and fund only the best possible technologies and solutions. When Boeing runs to congress and cuts NASA's funding for new orbiter research, they do good for no one.
In the life of sendmail, write-only cf files are a recent innovation.
There is still a huge amount of legacy material on modifying cfs. The perils of modifying the path to the local delivery agent (for example), or removing DaemonPortOptions (which has greater risks) are easily mitigated.
Please approach the subject with a more evenhanded point of view, and be aware of the historical perspective.
I am a U.S. Citizen, and I would like nothing more than to see all members of the RIAA and the MPAA dry up and blow away. I am certainly not alone in this view.
I look upon the recent financial woes of many of these corporations with schadenfreude - pleasure at the woes of another.
These organizations are making the US into a police state. They have orchistrated a coordinated attack upon our Bill of Rights, and they make a mockery of copyright law.
That they are suffering now is no great surprise. May it continue.
I hate rc.conf. RedHat's "service httpd restart" is much easier than digging the options up out of the/etc/rc* files.
In fact, I really miss the SysV init scripts. OpenBSD should have a switch you can throw that forces init to follow a SysV-style behavior. This would let the userbase decide how OpenBSD develops.
bzip2 is not part of the base distribution. RedHat converted to bzip2 some time ago, while OpenBSD still relies on gzip. This a) wastes download bandwidth, and b) wastes space on the OpenBSD cds. I probably shouldn't complain, but releases for some architectures aren't on the cds at all (alpha), and some have incomplete package collections (sparc). This is a stupid waste, forced by tradition alone. OpenBSD 4, when it comes out, should use the best compression available, no compromises.
A DVD distribution might be nice.
Mozilla doesn't work. Gimp is available, but there is no complete genome port. I hate konq.
(pd)ksh is not configured very well; everything is configured for (t)csh. There was no/etc/profile (if I remember correctly). It is sad to see OpenBSD and Apple using such a discredited shell. csh has profound problems as a scripting shell, and people should use the same shell for scripting that they use interactively.
sendmail.cf is not sufficiently flexible. Under RH6.2, I was able to use a mailertable entry of the form "pickydomain.net SMTP:[mail.myisp.com]" when a remote MTA insisted that I relay through my ISP. While I am no sendmail expert, this doesn't seem to be (easily) possible under the OpenBSD configuration. I copied RedHat's sendmail.cf to restore this functionality.
Some of my pcmcia network configuration was pure hell.
pkg_add should do more with PKG_PATH. Part of the installation should find the mirror with the best ping time/hop count/bandwidth, and install this mirror in/etc/PKG_PATH as the default source for all future installs. root should be able to trigger such a scan at any time.
The installer is a bit spartan, and also a bit confusing. OpenBSD's installer works in three phases: format file systems, untar the base operating system, configure the boot manager and sundry tasks. These sections should be clearly labeled, and something should be done to make them easier to understand (esp the "p m" option in creating the disklabel - "m" should be the default). A GUI instaler isn't really necessary, but something with curses would be better than what is currently used.
The documentation for the default NAT configuration (SOHO) ends up denying all connectivity to the OpenBSD router (RedHat does this also). When I want to turn on the POP server for a few minutes, I don't want to hassle with pf.
What is the sense in bundling a POP server in the base operating system, but not bundling an IMAP server?
Slashdot used to have a public link for this.
on
Absolute OpenBSD
·
· Score: 1
Slashdot used to maintain a link on the front page saying "Order your Amazon books here so that I can get a kickback" (or something similar). It was removed sometime around the Andover buyout.
Well, not that easily fixed. To be safe, any program dynamically-linking to libc had to be restarted, and any statically-linked binaries had to be recompiled (especially the NFS mountd). This at least should entail single user mode, if not a reboot.
Still, on the whole, (much) less work has to be done to keep OpenBSD patched than RedHat (especially when you add in the time for the up2date surveys).
Comparing the errata pages is sort of an apples-to-oranges situation, granted, but OpenBSD's record on 3.3 is impressive, none the less.
Correction: OpenBSD has had one libc errata. They have had no kernel errata. And yes, RedHat has issued a glibc errata in this time, too.
It is also not quite so misleading; OpenBSD bundles Apache, SSH, and Sendmail into the Base Operating System; RedHat does not. RedHat has also issued errata on OpenSSL, Apache (twice), and OpenSSH; none of these errata were necessary on OpenBSD. OpenBSD's errata page covers more, just as the OS does.
Why is BSD useful?
on
Absolute OpenBSD
·
· Score: 3, Informative
Let me count the ways...
Here is a list of the RedHat 9 errata. Here is the list of OpenBSD 3.3 errata. Notice a slight difference in the number of errata that have been issued between these distributions?
I may be generalizing, but when you need hardware compatibility, go with Linux; when you require security, go with OpenBSD.
If you run OpenBSD, you will immediately notice your Apache process:
httpd: parent [chroot/var/www] (httpd)
AFAIK, OpenBSD is the only UNIX(like) distribution with chroot Apache out of the box.
OpenBSD comes with the spamd daemon/system, which pulls information from spews.org and links it into the local pf routing, pointing spammers at a resource-consuming tarpit at little cost to you.
More importantly, this software is bound by the BSD license, which gives you much more freedom to do with the code as you will. Apple Mac OS X probably couldn't have been done with a GPL system (excepting gcc), for example. While the GPL is fantastic in that it will eventually destroy Microsoft, if you truly love software freedom, you will prefer BSD.
All the BSDs continue the legacy of the CSRG at UCB. Each (major version of) BSD is worth preserving for historical purposes alone.
There are quite a few things that I don't like about OpenBSD, but I've learned to live with them.
Consider working in a regional IT center of a large company. If you are accomplished and you attach yourself to a critical process, you can set yourself up as a small "fiefdom" and be mostly left alone.
RedHat has done some very great things for the Linux community. The GPL of the QT license is all due to them, and they were the only gnome player for the longest time.
This lawsuit with SCO is potentially another feather in their cap.
However...
It is pretty easy to show that RedHat doesn't care much about the Linux hobbiest community.
up2date surveys/support cuts for old releases
mp3
exiting the boxed set market
no reiserfs/xfs
no lvm
many Advanced Server features not backported to RH9
RedHat has to survive, granted, and that means money. Perhaps RedHat thinks that it used to waste money on a hobbiest market, but it is those users that have brought RedHat into the enterprise.
Memory scalability. x86 is 32-bit. UltraSPARC is 64-bit, and can address magnitudes more memory. Intel is dragging its feet on 64-bit x86 while pushing Itanium, while AMD pushes x86-64 extensions incompatible with anything produced by Intel. This situation is a mess; if your process needs >2gb of memory, you will have to leave Intel x86.
SMP Scalability. I believe that Unisys has scaled x86 to 64 cpus max. A Sun E15k scales to 100-120, and a LOT more people run >32 Sparc cpus in a box than Intel.
OS Scalability. Solaris partitions are much more extensive and powerful than VMWare, and this is only going to improve in Solaris 10, which runs one OS, but multiple, separate OS environments.
I/O Scalability. Solaris disk I/O is the best in the business. Your 33MHz PCI bus is old and slow by modern standards, and is now fragmenting into PCIX and PCI Express.
Reliability. Hot-swap components, shut down and locked out by the OS, have been part of Solaris servers for some time.
Standards compliance. Solaris is a UNIX98 OS. I don't think that there are any Intel UNIXen that meet this standard.
Processor megahertz is not everything, but Sun does need to ramp up the clock sometime soon.
Personally, I think that Sun should stop fighting and build an E15k with Opteron chips. AMD is also in great trouble... perhaps a Sun buyout would be beneficial to both.
Microsoft's last remaining stranglehold on the business desktop is the outlook/exchange integration. The protocol that is used is closed. Ximian Evolution advertises that it can function with exchange, but (AFAIK) it does so by faking an HTTP session with IIS (and you must enable web email logins for Evolution to work).
Windows is entrenched in well over 90% of desktop installations. While OpenOffice is giving the business community some serious ammo in license negotiations with Microsoft, very few (none?) of the Fortune 500 are dumping Office (let alone Win32) - yet.
You may assert that any business has the ability to choose a Linux desktop, but this choice will only be made when Linux is both far cheaper and far more user-friendly than now - to justify the pain that an enterprise would feel going cold turkey from the MS crack addiction.
The market is sick of MS tactics - it is interesting to watch MS being contained in the cell phone industry, for example. But that doesn't mean that MS could disappear tomorrow, or that Linux is a completely viable desktop, and that certainly doesn't mean that the MS-Linux playing field is either level or fair.
Slashdot Mirror
No, I didn't take your comments in too negative a light.
There are some activities where a centralized approach makes sense. Napster was technically a better system than Kazaa.
However, in this case as in Napster, any centralization will be the focus of an attack. A blacklist system should be designed to make an effective attack extremely difficult and expensive. I can't think of a better way to accomplish this, although it brings to mind the dreaded "p2p" word.
SQL Server will never run on any version of UNIX. AFAIK, there aren't even (MS-supported) SQL Server client libraries for non-windows platforms. I realize that FreeTDS is available, but such a library would never be used in a highly critical sector.
If you have to integrate multiple platforms, you cannot use SQL server. Closest similar product is Sybase ASE, but Microsoft broke Sybase compatibility on purpose.
I think if people understood how irrationally obstinate SQL Server's platform dependence was, they would look elsewhere. I hope that this attitude holds them below 10% penetration - it certainly has up to now. They are a bit player.
In any case, there is a cheap, new version of DB2 out for $500/copy.
Actually, the system that I just described bears far more similarity to NNTP than a Napster or Kazaa protocol.
NNTP predates most p2p by a long time, and seems to have been resilient in maintaining basic connectivity.
Do you have a better idea?
I can easily see web content filtering going the same way eventually.
It is good to see IBM ending the habit of charging extra for the C compiler. AIX hasn't bundled the compiler since 3.2.5.
I understand that the ESA is testing a new orbiter. Why is there no broad collaboration by NASA in this effort?
Could it be that Boeing and other major contractors behind the shuttle continually flex their political muscle to keep the shuttle program alive because of the large maintenance contracts?
NASA should find a way to shove the contractors in the closet and fund only the best possible technologies and solutions. When Boeing runs to congress and cuts NASA's funding for new orbiter research, they do good for no one.
In the life of sendmail, write-only cf files are a recent innovation.
There is still a huge amount of legacy material on modifying cfs. The perils of modifying the path to the local delivery agent (for example), or removing DaemonPortOptions (which has greater risks) are easily mitigated.
Please approach the subject with a more evenhanded point of view, and be aware of the historical perspective.
I am a U.S. Citizen, and I would like nothing more than to see all members of the RIAA and the MPAA dry up and blow away. I am certainly not alone in this view.
I look upon the recent financial woes of many of these corporations with schadenfreude - pleasure at the woes of another.
These organizations are making the US into a police state. They have orchistrated a coordinated attack upon our Bill of Rights, and they make a mockery of copyright law.
That they are suffering now is no great surprise. May it continue.
On one platform, you can:
Thorough study of Mac OS X can land a student a $100k+ job. Thorough study of Microsoft platforms gets a student an MCSE and $8.50/hr.
School administrators, do not cripple your students with Microsoft products.
Slashdot used to maintain a link on the front page saying "Order your Amazon books here so that I can get a kickback" (or something similar). It was removed sometime around the Andover buyout.
Well, not that easily fixed. To be safe, any program dynamically-linking to libc had to be restarted, and any statically-linked binaries had to be recompiled (especially the NFS mountd). This at least should entail single user mode, if not a reboot.
Still, on the whole, (much) less work has to be done to keep OpenBSD patched than RedHat (especially when you add in the time for the up2date surveys).
Comparing the errata pages is sort of an apples-to-oranges situation, granted, but OpenBSD's record on 3.3 is impressive, none the less.
Correction: OpenBSD has had one libc errata. They have had no kernel errata. And yes, RedHat has issued a glibc errata in this time, too.
It is also not quite so misleading; OpenBSD bundles Apache, SSH, and Sendmail into the Base Operating System; RedHat does not. RedHat has also issued errata on OpenSSL, Apache (twice), and OpenSSH; none of these errata were necessary on OpenBSD. OpenBSD's errata page covers more, just as the OS does.
Let me count the ways...
I may be generalizing, but when you need hardware compatibility, go with Linux; when you require security, go with OpenBSD.
AFAIK, OpenBSD is the only UNIX(like) distribution with chroot Apache out of the box.
There are quite a few things that I don't like about OpenBSD, but I've learned to live with them.
...because of the exposure to dangerous/carcinogenic solvents and substances.
Last time I heard, the biggest IBM mainframe was based on a 16-way, 700MHz system (with another 16 redundant processors).
These systems woefully underperform modern UNIX, and IBM no longer publishes benchmarks.
IBM recently won two rounds of TPC benchmarks with DB2 on AIX - not z/os.
Consider working in a regional IT center of a large company. If you are accomplished and you attach yourself to a critical process, you can set yourself up as a small "fiefdom" and be mostly left alone.
Is this an old 601/604 model? Or is it a G3?
I would assume that an application like this requires SIMD fp operations.
Is it 32 or 64 bit?
RedHat could slam them around in court a bit to drive down their stock price, then pick them up for spare change.
RedHat has done some very great things for the Linux community. The GPL of the QT license is all due to them, and they were the only gnome player for the longest time.
This lawsuit with SCO is potentially another feather in their cap.
However...
It is pretty easy to show that RedHat doesn't care much about the Linux hobbiest community.
RedHat has to survive, granted, and that means money. Perhaps RedHat thinks that it used to waste money on a hobbiest market, but it is those users that have brought RedHat into the enterprise.
Now I bring SUSE and OpenBSD.
Processor megahertz is not everything, but Sun does need to ramp up the clock sometime soon.
Personally, I think that Sun should stop fighting and build an E15k with Opteron chips. AMD is also in great trouble... perhaps a Sun buyout would be beneficial to both.
Microsoft's last remaining stranglehold on the business desktop is the outlook/exchange integration. The protocol that is used is closed. Ximian Evolution advertises that it can function with exchange, but (AFAIK) it does so by faking an HTTP session with IIS (and you must enable web email logins for Evolution to work).
Windows is entrenched in well over 90% of desktop installations. While OpenOffice is giving the business community some serious ammo in license negotiations with Microsoft, very few (none?) of the Fortune 500 are dumping Office (let alone Win32) - yet.
You may assert that any business has the ability to choose a Linux desktop, but this choice will only be made when Linux is both far cheaper and far more user-friendly than now - to justify the pain that an enterprise would feel going cold turkey from the MS crack addiction.
The market is sick of MS tactics - it is interesting to watch MS being contained in the cell phone industry, for example. But that doesn't mean that MS could disappear tomorrow, or that Linux is a completely viable desktop, and that certainly doesn't mean that the MS-Linux playing field is either level or fair.
Didn't The Open Group do an entire UNIX implementation (the only implementation of which was Digital OSF/1|UNIX|Tru64)?
If so, how much of this could they open? Anything useful in it?
MS integrated OLAP software into SQL server without permission, and a court awarded triple damages against the SQL Server userbase.
Read it and weep. I guess we should all just stick to stone tablets.
If you are on the list, go out and buy a wireless router, and attach it via NAT/ICS to your main network connection.
Pay cash for the router, and try to buy an older model.
How can the RIAA prove that it was you?