I think it's only a matter of time before the official Linux kernel development switches to a "core group" model working against a private source tree using version management software rather than a straight hierarchy. Core developers can check in, everyone else sends patches. When it happens it will be quick and the scalability benefits will be obvious.
I'm not saying this as flamebait or to say so-and-so is wrong. I have a great deal of respect for Linus Torvalds but his brain-child has outgrown him and at some point he needs to let it go outside the house.
The X11 and the BSD licenses are a mixed bag. On one hand there are amazing success stories like well, XFree86 and the BSDs. On the other hand there are less successful scenarios like WINE where you see lots of proprietary forks, all closed from each other, all providing a disincentive for someone to redundantly code a free version.
The future of Mono as free software appears to be troubled (even more than it already was). It may make Ximian and Intel a buck or two, but I wonder if it will bring any benefit to the free sofware community.
I would have preferred an LGPL license. Dropping the LGPL, which is really quite tame, makes me suspicious of Ximian/Intel's motives.
I've used both Ximian's redcarpet updater and RHN. Ximian's client is very slick looking, and I like how you can subscribe to different channels. I also like you there's an option to install RPMs from a directory- this makes Redcarpet useful as a general RPM frontend.
Red Hat Network doesn't look as elegent, but it has more functionality overall. RHN gives you emails of critical updates and errata tailored to your actual systems. It gives you a single point of management for multiple computers. One thing that I tried yesterday was to schedule the install of new RPM packages on one of my tower systems from my laptop. It worked great! Also, I love how you can exclude packages from being upgraded. It's very annoying to have to click on every individual update manually in red carpet just because if you hit "update all" it will update a package you want left alone (usually replacing a more up to date version with a downgraded ximian version).
I'm used to periodically checking for updates manually, and then pushing them down to each system as needed. This saves a lot of time if you can spare 60$ a year for each additional system (you get one system free).
For two disks, you should get the best results with both disks configured as masters on two different IDE buses.
If you're not seeing that, I'd check that you have the correct drivers/optimizations for your IDE chipset enabled. You also might want to check IRQ allocation to make sure there's no strange conflicts . Check your windows (NT/2000) event log to make sure there's no strange IDE timeouts indicating hardware issues. If you still see the problem you should try your test on a different hardware platform (motherboard/controller combo).
From your description, however, you might want to go with a raid technology such as RAID 1, RAID 5, or raid 1+0. It will offer much better redundancy and possibly improved performance.
I know that as a Linux enthusiast and part time advocate, I will not recommend a TimeWarner/AOL distribution.
Who could?
Who could give money to an organization that actively props up UTICA, DMCA, and a whole network of proprietary standards full of ads, user tracking devices, content control schemes?
For Alan, and other people who make the effort to pay attention to issues beyond their paycheck, there's more to life than getting paid.
Think about what TW/AOL does. Think especially about their leading role in content control legislation, UTICA, and the DMCA. Think about how they've compromised all the software they've absorbed with unsolicited advertising, user tracking, shopping portals and the like.
It's more difficult to stand up for what you think is right than wallow in apathy. We should all support Alan and those like him for refusing to support an organization that acts directly against his core beliefs.
You're asking the wrong question. The question that you want to ask is "If RedHat was bought by TimeWarner/AOL, wouldn't that be good?"
The answer is HELL NO.
AOL/TimeWarner is already pushing content control schemes across the board. These schemes are in opposition to RedHat's vision for Linux.
Look at what TW/AOL's has been doing to their owned companies have been doing since the merger-
Some of their companies have been forced out of their market not because they weren't making a profit, but because they weren't making *enough* of a profit. Across the board of TW/AOL you see the dumbing down of quality in the name of greed, you see the explicit compromise of journalism to advertisers, and ads, tracking, and spam all over their tech products.
AOL/TW is a company with no soul, a company that has compromised every product they had of any worth. What future do you see for RedHat in this environment? I see the best known Linux brand installing AOL, AOLIM, and all sorts of shopping junk, advertisements, spam, and tracking bugs onto the default desktop.
The GPL doesn't protect about that. They want to subvert the RedHat brand to push their crap.. it's has nothing to do with the technical qualities of the distribution, which will soon suffer.
I have a friend who works at a company owned by AOL/TW. Since the merger they've been forced to create products more suitable for walmart than their target market. They had their groupware replaced with a custom, more restrictive version of AOL mail, which comes complete with SPAM from AOL from day 1, and *requires* you to be logged in to AOL/IM in order to check your mail.
There's reason AOL would want to buy RedHat instead of partnering with them is if they wanted to subvert the brand. Wake Up.
As someone who is familiar with some of the inner workings of post-merger AOL/TW I don't see any honorable future for RedHat should they be bought.
So far, AOL/TW's pattern since the merger has been to cut spending and investment in their held companies, demand 10% growth from them, and should the small company fail to deliver they are mercilessly cut down until they can or they are gone.
Red Hat will not fit into this kind of mold. Red Hat will be forced out of their current businesses and cease to operate independently. They will turn into yet another AOL platform or yet another AOL lure. In a few years AOL/RH will be pushing content-control mechanisms on default installs of their distribution.
Perhaps RH should never have gone public and exposed themselves to something like this. An AOL-RH will not represent the current Linux community like RH does today. Most likely we'll have to find something else.
I don't understand why AOL/TW won't simply partner with redhat to get the services they want. I have a feeling that they're after the default install of a leading brand name distribution, and that just stinks.
Yes, I know, but how many people have RTF viewers built in to their email client? It's not as supported as plaintext or HTML. I'm not saying that RTF is bad, but it's not the best solution.
That misses the point.. Wordpad is certainly inferior to plaintext and html as far as being able to read it on other platforms easily. Start emailing people wordpad files and see how well that takes off.
What would be better is if an easy to use, cross-platform editor was bundled with windows. Something with the most basic functions including search, replace, simple formatting, and the ability to deal with files larger than 32k (unlike most versions of notepad).
It could be a text editor or a lightweight WSIWYG HTML editor. Either way, if it was on the system by default it would be very useful.
For most tasks, notepad is too little and Word is too much.
A little off-topic, but many people in this story have written comments about authenticating their print server against "LDAP". All I know about LDAP is that it's a "Lightweight Directory Access Protocol".
What are people using for LDAP server software? Are there any Free solutions with graphical management tools? What platforms are available?
If someone from Zap is reading this, what I would like is for it to support the ogg/vorbis audio format, use DivX for video ripping, and have some kind of "expert-level" access to the underlying OS for people that want to tweak it. Oh, and get the price down under 800$.
Don't proselytize. Get her an mid to upper range Mac, convince her that she will need to spend more but it is worth it.
If you go with MacOSX you can get her to try other OS's and applications later down the road.
For things like font handling, printing, and media creation *NO* major operating system beats the quality of apple's default tools. There's plenty of third party applications as well, just ask Adobe. Do not underestimate the power of being able to drag applications and documents all around the hard drive and still have them work flawlessly. Do not underestimate how nice it is to have application directories that are not full of confusingly named support files.
If your friend wants to be an artist, not a tech enthusiast she will not appreciate having to do extra administration work on operating systems that do not cater to her community.
I was a consultant for a media house that had a mix of windows, macs, (and when I was done) linux servers. The best people there regarded the windows platform as a necessary evil for doing business with non-artist customers, but never their first choice for a job.
I wish you luck.
It's been a year since I've researched the subject, but some of this info still may be relevant. If not, I'm sure I'll be moderated down and/or corrected:)
LIDS and Snort do very different things. LIDS is more for host-based security. It is primarily used for locking down the kernel. For example, adding additional layers of security to prevent unauthorized kernel module loading, file access, etc. It foils common rootkits and can be used to make a hardened machine. The downside is that it works at a very low level. You have to patch your kernel to get it to work, and the LIDS package lag behind the linus tree. The configuration interface at the time I looked at it was in flux and poorly documented. It might be better now, but it looked like it took a lot of effort to customize a configuration to meet your particular needs.
Snort is a whole different story. It is used to report suspicious network activity, such as portscans, web server attacks, ftp overflow attacks, etc. The snort scanning engine is quite sophisticated and easily customizable by rules files. It appears to be every bit as effective as commercial equivalents if not better. The downside is that the reporting is very do-it-yourself. If you want to get something more than spammy SYSLOG alerts, you have to roll your own reporting/alert/reaction tool. To be fair there are lots of hooks and database-backend support for this, but it doesn't come with the base package. Perhaps someone will reply with a link to a third-party add on that fills this gap.
Not to be contrary, but the weaknesses you describe are minor.
You can make a list with announce-only properties by configuring it as moderated, and make the person/accounts who sends the announcements the moderator.
Your points #2 and #3 are really the same argument, that it's better to trade the security for convenience. Obviously, that's debatable and many people will see it as a feature rather than a bug. In practice it is not hard to enter a password in order to unsubscribe.
Mailman works for me. It doesn't have as many features as listserv, but it's easy to set up, easy to customize, and easy to administer. It is so easy that people can manage their own lists, which is a good thing for those of us who dislike being interrupted for simple adds/changes. Of course, GNU Mailman is free software. I think it's a no brainer for most small-midsized lists.
This isn't a terribly insightful comment, but the technical differences between commercial ssh and openssh are minor, even trivial.
If spending money for support and proactive updates is easier for company than having a your current IT staff RTFM and monitor security-related mailing lists, then go with commercial ssh.
Remember too that in almost all cases openssh and commercial ssh can interoperate. So, you could buy commercial ssh on a few machines until you're confortable with using it, and then implement future installations using openssh.
I believe that it's a good thing to have a security policy. However, there are lots of vulnerability forums that are much larger in scope than Mozilla.
What does it matter what Mozilla's official vulnerability warning policy is when you can get the info from bugtraq etc?
The trick you missed with the zombies is that you can simply walk up to them with a sword, tilt back far enough so you see the top half of their body, swing and decapitate in one go.
No jumping, no fancy aiming.. It decapitates 70% of the time in one hit.
I know the article is more about performance, but I have to point out that the reason I'm using XFS right now is scalability. It is *VERY NICE* to be able to create files greater than 2GB on a 32bit CPU, among other things.
As for performance.. the test was for "lots of little tiny files". ReiserFS is supposed to be the champ at this. XFS is tuned for larger sizes on larger sized filesystems. It has support for guaranteed-bandwidth (realtime support), ACLs, and as the ginsu-knife salesman says- "and much, much more!".
One other thing worth mentioning is that XFS comes with mature userland tools.
Slashdot Mirror
I think it's only a matter of time before the official Linux kernel development switches to a "core group" model working against a private source tree using version management software rather than a straight hierarchy. Core developers can check in, everyone else sends patches. When it happens it will be quick and the scalability benefits will be obvious.
I'm not saying this as flamebait or to say so-and-so is wrong. I have a great deal of respect for Linus Torvalds but his brain-child has outgrown him and at some point he needs to let it go outside the house.
The X11 and the BSD licenses are a mixed bag. On one hand there are amazing success stories like well, XFree86 and the BSDs. On the other hand there are less successful scenarios like WINE where you see lots of proprietary forks, all closed from each other, all providing a disincentive for someone to redundantly code a free version.
The future of Mono as free software appears to be troubled (even more than it already was). It may make Ximian and Intel a buck or two, but I wonder if it will bring any benefit to the free sofware community.
I would have preferred an LGPL license. Dropping the LGPL, which is really quite tame, makes me suspicious of Ximian/Intel's motives.
I've used both Ximian's redcarpet updater and RHN. Ximian's client is very slick looking, and I like how you can subscribe to different channels. I also like you there's an option to install RPMs from a directory- this makes Redcarpet useful as a general RPM frontend.
Red Hat Network doesn't look as elegent, but it has more functionality overall. RHN gives you emails of critical updates and errata tailored to your actual systems. It gives you a single point of management for multiple computers. One thing that I tried yesterday was to schedule the install of new RPM packages on one of my tower systems from my laptop. It worked great! Also, I love how you can exclude packages from being upgraded. It's very annoying to have to click on every individual update manually in red carpet just because if you hit "update all" it will update a package you want left alone (usually replacing a more up to date version with a downgraded ximian version).
I'm used to periodically checking for updates manually, and then pushing them down to each system as needed. This saves a lot of time if you can spare 60$ a year for each additional system (you get one system free).
No offense, but is there a set of icons/theme that make it look like Gnome?
I think that KDE's technologies with Gnome's look would be attractive to many people.
> But where is Borland to go now? As I see it, two
> areas they need to focus on (just my opinion):
How about dropping the price down from $3,000 for the version of the product with the features that you mention?
I mean, *really*. It's a great product, but not that great. I haven't seen a markup like that since I bought software for the AS400.
Your results sound strange to me.
For two disks, you should get the best results with both disks configured as masters on two different IDE buses.
If you're not seeing that, I'd check that you have the correct drivers/optimizations for your IDE chipset enabled. You also might want to check IRQ allocation to make sure there's no strange conflicts . Check your windows (NT/2000) event log to make sure there's no strange IDE timeouts indicating hardware issues. If you still see the problem you should try your test on a different hardware platform (motherboard/controller combo).
From your description, however, you might want to go with a raid technology such as RAID 1, RAID 5, or raid 1+0. It will offer much better redundancy and possibly improved performance.
I know that as a Linux enthusiast and part time advocate, I will not recommend a TimeWarner/AOL distribution.
Who could?
Who could give money to an organization that actively props up UTICA, DMCA, and a whole network of proprietary standards full of ads, user tracking devices, content control schemes?
Only someone who didn't care, and shame on you.
For Alan, and other people who make the effort to pay attention to issues beyond their paycheck, there's more to life than getting paid.
Think about what TW/AOL does. Think especially about their leading role in content control legislation, UTICA, and the DMCA. Think about how they've compromised all the software they've absorbed with unsolicited advertising, user tracking, shopping portals and the like.
It's more difficult to stand up for what you think is right than wallow in apathy. We should all support Alan and those like him for refusing to support an organization that acts directly against his core beliefs.
You're asking the wrong question. The question that you want to ask is "If RedHat was bought by TimeWarner/AOL, wouldn't that be good?"
The answer is HELL NO.
AOL/TimeWarner is already pushing content control schemes across the board. These schemes are in opposition to RedHat's vision for Linux.
Look at what TW/AOL's has been doing to their owned companies have been doing since the merger-
Some of their companies have been forced out of their market not because they weren't making a profit, but because they weren't making *enough* of a profit. Across the board of TW/AOL you see the dumbing down of quality in the name of greed, you see the explicit compromise of journalism to advertisers, and ads, tracking, and spam all over their tech products.
AOL/TW is a company with no soul, a company that has compromised every product they had of any worth. What future do you see for RedHat in this environment? I see the best known Linux brand installing AOL, AOLIM, and all sorts of shopping junk, advertisements, spam, and tracking bugs onto the default desktop.
The GPL doesn't protect about that. They want to subvert the RedHat brand to push their crap.. it's has nothing to do with the technical qualities of the distribution, which will soon suffer.
I have a friend who works at a company owned by AOL/TW. Since the merger they've been forced to create products more suitable for walmart than their target market. They had their groupware replaced with a custom, more restrictive version of AOL mail, which comes complete with SPAM from AOL from day 1, and *requires* you to be logged in to AOL/IM in order to check your mail.
There's reason AOL would want to buy RedHat instead of partnering with them is if they wanted to subvert the brand. Wake Up.
As someone who is familiar with some of the inner workings of post-merger AOL/TW I don't see any honorable future for RedHat should they be bought.
So far, AOL/TW's pattern since the merger has been to cut spending and investment in their held companies, demand 10% growth from them, and should the small company fail to deliver they are mercilessly cut down until they can or they are gone.
Red Hat will not fit into this kind of mold. Red Hat will be forced out of their current businesses and cease to operate independently. They will turn into yet another AOL platform or yet another AOL lure. In a few years AOL/RH will be pushing content-control mechanisms on default installs of their distribution.
Perhaps RH should never have gone public and exposed themselves to something like this. An AOL-RH will not represent the current Linux community like RH does today. Most likely we'll have to find something else.
I don't understand why AOL/TW won't simply partner with redhat to get the services they want. I have a feeling that they're after the default install of a leading brand name distribution, and that just stinks.
Yes, I know, but how many people have RTF viewers built in to their email client? It's not as supported as plaintext or HTML. I'm not saying that RTF is bad, but it's not the best solution.
That misses the point.. Wordpad is certainly inferior to plaintext and html as far as being able to read it on other platforms easily. Start emailing people wordpad files and see how well that takes off.
What would be better is if an easy to use, cross-platform editor was bundled with windows. Something with the most basic functions including search, replace, simple formatting, and the ability to deal with files larger than 32k (unlike most versions of notepad).
It could be a text editor or a lightweight WSIWYG HTML editor. Either way, if it was on the system by default it would be very useful.
For most tasks, notepad is too little and Word is too much.
Forget vim.. All someone has to do is bundle a better text editor than notepad with windows by default and people will be happy with that.
This might be something that could be done pending the outcome of the MS trial.
A little off-topic, but many people in this story have written comments about authenticating their print server against "LDAP". All I know about LDAP is that it's a "Lightweight Directory Access Protocol".
What are people using for LDAP server software? Are there any Free solutions with graphical management tools? What platforms are available?
Thanks in advance
If someone from Zap is reading this, what I would like is for it to support the ogg/vorbis audio format, use DivX for video ripping, and have some kind of "expert-level" access to the underlying OS for people that want to tweak it. Oh, and get the price down under 800$.
Don't proselytize. Get her an mid to upper range Mac, convince her that she will need to spend more but it is worth it.
If you go with MacOSX you can get her to try other OS's and applications later down the road.
For things like font handling, printing, and media creation *NO* major operating system beats the quality of apple's default tools. There's plenty of third party applications as well, just ask Adobe. Do not underestimate the power of being able to drag applications and documents all around the hard drive and still have them work flawlessly. Do not underestimate how nice it is to have application directories that are not full of confusingly named support files.
If your friend wants to be an artist, not a tech enthusiast she will not appreciate having to do extra administration work on operating systems that do not cater to her community.
I was a consultant for a media house that had a mix of windows, macs, (and when I was done) linux servers. The best people there regarded the windows platform as a necessary evil for doing business with non-artist customers, but never their first choice for a job.
I wish you luck.
It's been a year since I've researched the subject, but some of this info still may be relevant. If not, I'm sure I'll be moderated down and/or corrected :)
LIDS and Snort do very different things. LIDS is more for host-based security. It is primarily used for locking down the kernel. For example, adding additional layers of security to prevent unauthorized kernel module loading, file access, etc. It foils common rootkits and can be used to make a hardened machine. The downside is that it works at a very low level. You have to patch your kernel to get it to work, and the LIDS package lag behind the linus tree. The configuration interface at the time I looked at it was in flux and poorly documented. It might be better now, but it looked like it took a lot of effort to customize a configuration to meet your particular needs.
Snort is a whole different story. It is used to report suspicious network activity, such as portscans, web server attacks, ftp overflow attacks, etc. The snort scanning engine is quite sophisticated and easily customizable by rules files. It appears to be every bit as effective as commercial equivalents if not better. The downside is that the reporting is very do-it-yourself. If you want to get something more than spammy SYSLOG alerts, you have to roll your own reporting/alert/reaction tool. To be fair there are lots of hooks and database-backend support for this, but it doesn't come with the base package. Perhaps someone will reply with a link to a third-party add on that fills this gap.
If you read the fine print on Ani's "Righteous Babe Records" CDs you will see that they are quite permissive with regard to copying.
It is highly unlikely that these CDs would be part of any CD-based copy control scheme.
Not to be contrary, but the weaknesses you describe are minor.
You can make a list with announce-only properties by configuring it as moderated, and make the person/accounts who sends the announcements the moderator.
Your points #2 and #3 are really the same argument, that it's better to trade the security for convenience. Obviously, that's debatable and many people will see it as a feature rather than a bug. In practice it is not hard to enter a password in order to unsubscribe.
Mailman works for me. It doesn't have as many features as listserv, but it's easy to set up, easy to customize, and easy to administer. It is so easy that people can manage their own lists, which is a good thing for those of us who dislike being interrupted for simple adds/changes. Of course, GNU Mailman is free software. I think it's a no brainer for most small-midsized lists.
This isn't a terribly insightful comment, but the technical differences between commercial ssh and openssh are minor, even trivial.
If spending money for support and proactive updates is easier for company than having a your current IT staff RTFM and monitor security-related mailing lists, then go with commercial ssh.
Remember too that in almost all cases openssh and commercial ssh can interoperate. So, you could buy commercial ssh on a few machines until you're confortable with using it, and then implement future installations using openssh.
-D
I believe that it's a good thing to have a security policy. However, there are lots of vulnerability forums that are much larger in scope than Mozilla.
What does it matter what Mozilla's official vulnerability warning policy is when you can get the info from bugtraq etc?
The trick you missed with the zombies is that you can simply walk up to them with a sword, tilt back far enough so you see the top half of their body, swing and decapitate in one go.
No jumping, no fancy aiming.. It decapitates 70% of the time in one hit.
I know the article is more about performance, but I have to point out that the reason I'm using XFS right now is scalability. It is *VERY NICE* to be able to create files greater than 2GB on a 32bit CPU, among other things.
As for performance.. the test was for "lots of little tiny files". ReiserFS is supposed to be the champ at this. XFS is tuned for larger sizes on larger sized filesystems. It has support for guaranteed-bandwidth (realtime support), ACLs, and as the ginsu-knife salesman says- "and much, much more!".
One other thing worth mentioning is that XFS comes with mature userland tools.
Not to be picky but the gauls had a celtic culture related to the better known celts in the british isles.
The romans had a tendancy of calling all non-romans barbarians. It's a bit too subjective a term.