Wow.. Debian *STABLE* releases contain older, more stable software.
So what if Debian's stable releases are long and far between? They still release much more often than Microsoft, Apple, and Sun do and with many more applications.
If you've ever ran Debian 'testing' or 'unstable', you would realize that NO other distribution or operating system is nearly as upto-date as Debian. It is only the 'stable' branch that is older, with good reason.
The GPL makes it impossible to link to non-GPL code. However, it is not illegal to distribute code that cannot be compiled.
The developers of MPlayer did not break the GPL by distributing source code. Users whom linked GPL code with non-GPL code broke the GPL.
Although it may not be 'ok' for users to make illegal binaries, the developers of the software are not to blame. The MPlayer developers simply made a note stating that it could not be compiled legally, which was responsible.
Because the Net2Phone software has the algorithm implimented in software. It is a patented algorithm and cannot be used in free software, hence free software must currently use a hardware-based solution.
The hardware apparently has some advantages over software-only implimentations, including the usage of a real telephone.
vonage.com will work for those without Microsoft Windows as it is a hardware-only solution, it connects to your hub/switch and 'just works'. The nice thing you get with GnomeMeeting is the ability to dial from the computer and the option to use the computer to speak or process the audio data.
The advantage with Quicknet vs Vontage.com is that Quicknet does not (currently?) have a monthly access fee. Vontage will charge you a monthly fee, for unlimited calling (or less for 500 domestic minutes).. Quicknet charges per-minute and is blind to national borders.
Sure, knowing and distributing ips might seem like a good idea.. but it would be a nightmare for usability. You may say, "but just put the IP in the address bar of a webbrowser"; however, if that is your argument than you must have forgotten about HTTP/1.1. HTTP/1.1, and other protocols such as SMTP allow multiple domain-names exist at the same IP address. If you do not pass the domainname AND the IP with these protocols, the servers will default to god-knows-what.
We would have been ok without dns, but currently available services would be difficult or impossible to replace. Rather than typing http://192.168.1.2/, you would have to type http://somedomain.com!192.168.1.2/ or other such abdomination (@ is not used as it is popularly used to designate HTTP (and sometimes FTP) authentication).
If we didn't use DNS and such hostname-based hacks didn't appear in HTTP and SMTP, we would have a much greater need for IPV6!
DNS is a good thing, dns servers must be repaired.
Just so you're aware, one firewire bus can only do 400Mbps. You might want more than 100Mbps (~12MB/s) per drive.:)
Personally, I'd go with buying an empty drive chassis with 5.25" drive bays.. then create a firewire backplane and use those ide drive caddies for holding the disks.
The only question is how many firewire controllers do you wish to have ?:)
At the cost of a little overhead, you can run MacOS under the MacOnLinux VM. Any PowerPC machine that can run linux can run MacOS 7-10.
If you don't think the fastest apple-branded machine is suitable.. how about a quad-cpu Power4 machine from IBM? They do have some powerful workstations, although the servers could be fine too as you could export the MacOnLinux display via X11 or it's (optional) internal VNC server. It supports the Linux framebuffer too if you don't want the overhead of an X11 server.
> This may work well for a school, but everybody > around here realizes that monitors almost always > out-live just about every other component. The iBook
Actually, I wouldn't buy an (i|e)Mac because I fear that the monitor would die leaving me with an unusable system!!
In my experience, monitors die before anything else. I went through 4 monitors in 2002... and 2 monitors so-far in 2003 (noting that it is still January!).. granted, 2/6 of those monitors were quite old (9 years)..
I've been finding that newer CRTs are less reliable than the older models. Is it sane that I have an 8month old monitor die within the same week as a 9 year old monitor? Something is obviously wrong with the manufacturing of that 8 month old monitor! Not to mention that the 8 month old monitor was a *replacement* for a monitor that was bought one year before it's death.
btw, environmental issues have been ruled out as the cause of death for these monitors.
- You do not need to buy apple to run MacOS. - There ARE PPC motherboards, abeit expensive... but that is because there isn't a big market for them. There is a board designed for Amiga OS4 that works fine with Linux as well as the IBM machines. The importance is that any PPC board that runs linux can run MacOS, quickly too. - OSX will not come to x86, are you nuts? Apple would not dare lose their superior architecture and try to fight Microsoft in a 1:1 battle. The best that happens from x86 OSX is that a few Windows users will dual-boot or use VMware. Apple doesn't make their money on software.
You are right about one thing, the Xserve is underpowered and over-expensive.
Except you don't find $1/GB on SCSI.. especially not SCA if you plan to hot-swap. I suspect it would be much cheaper to obtain Firewire->IDE adapters which would allow you to hot-swap.
Of course, Firewire1 isn't fast enough.. you would have to go with Firewire2 for any reasonable ammount of speed. 400Mbps
120GB * 9 = 1080GB (120GB drives are cheapest) $170 * 9 = $1530 ($170 for a 120GB drive and firewire2 enclosure)
Each Firewire2 bus can do 800Mbps, which is 100MBps.. you would want each drive doing at least 50Mbps, requiring 5 Firewire2 controllers (and associated costs) in your machine... assuming you can FIND Firewire2 controllers (I haven't been able to, other than what is builtin to the new Apple machines). You could go with slower speeds for less money, your choice.
Of course, as the original poster said.. you can just use IDE.
Imagine how many cows/chickens/turkeys/pigs/etc we wouldn't kill if we were happy just drinking their blood... sure, we would probably have to have a lot more of those animals in captivity at one time.. but the total number of animals required for any extended period of time would be much lower..
Humans need to evolve to requiring only blood and other disposible parts.. better to hurt animals than kill them. Better than vegatables too.. vegatarians think they're so much better because they don't kill animals... they never stop to think about the plants! Plants have feelings too.
Mozilla may not have the greatest share of the market, but it may be the best browser available. This is why Apple DOES NOT want Mozilla. Sounds crazy? Not really.
Jobs realizes that competition will create better software. It would certainly be possible for Mozilla to become so popular and 'standardized' on the Unix and MacOS operating systems that development of KHTML would slow down and eventually die. If you have a company behind KHTML like Apple while AOL is behind Mozilla, you can expect a war to brew.
Mozilla is a great browser, KHTML is not bad.. but unless they become more popular and gain more press, Microsoft won't bother to compete.. they won't have to.
If KHTML and Mozilla begin a new browser war, first.. new OSX users will be using KHTML, Linux/Unix geeks will be using either Mozilla or KHTML. Apple still does have a large userbase, using KHTML could really put a dent in Microsoft. KHTML's competition would make Mozilla better and more popular, even in Microsoft Windows.
Apple may have just sparked not only a browser war, but a rejuvination of computing without Microsoft. I won't be surprised to see 30-40% of the web using non-IE browsers within a year.
Using font-size: small should not post a significant problem with any browser if implimented properly. It is true that it may provide a smaller font than the user has requested of their browser; however, the browser should render it at a slightly smaller, yet readable size.
With the default settings of every browser I've tried (Opera5/6-mac, IE4.5/5.1-mac, Galeon/Mozilla-linux, Mozilla-mac, Netscape4-mac), font-size: small provides readable fonts..
Not that I terribly condone the practice of making the font smaller than the user's preferences.. I personally think that everyone should have their browser properly configured and let designers make their pages accessible; however, if you must make the font smaller.. font-size: small is the safest and most accessible-friendly way to change it.
Realizing the 'open source' (sort of) nature of Darwin, I've thought of doing this myself (for linux).. but just didn't have them time.
For those who think this is a dumb idea.. you're not seeing the bigger picture. Currently, MacOS X only runs on G3 and G4 apple machines designated by Apple unless you boot Linux and install Mac On Linux (similiar to Plex86 for PPC). If you can run OSX applications on Linux, NetBSD, etc.. then you wouldn't need the overhead of a VM. Your RS6000 could run OSX with minimum overhead. From NetBSD you could run from the same machine, binaries compiled for: OSX, Linux, FreeBSD, and NetBSD.
Personally, I don't like OSX; However, I wouldn't mind being able to run natively on Linux: Internet Explorer, Quarx Express, etc.. Not to mention that Quartz is a pretty neat looking, X11 compatable (with Xfree86-rootless) windowing system.
This is more important for Linux than NetBSD, only because NetBSD currently has poor support for devices on many PowerPC machines (compared to Linux), especially laptops.
I write my pages with valid XHTML and CSS2. The pages will render perfectly in Mozilla and with slight tweaks, Konqueror and Opera. IE4.5 and IE5 will generally render quite well; however, IE6 is (almost?) IMPOSSIBLE to get to work.. the browser is just plain broken. It still doesn't properly render my pages.
Microsoft actually had better CSS support in IE5.x than it does in 6.
> "I've also been thinking that a filesystem such as > NTFS or ReiserFS that allows attaching huge > ammounts of metadata, or small amounts of metadata > to any file would be important. For instance, my > 4096x2048 digital photograph of the grand canyon > (big file), should still be able to have a > thumbnail (say about 128 KB) attached as metadata."
Uh, you mean like ADS in NTFS? Some thumbnail programs use it for keeping the thumbnails attached. The data is automatically discarded by NT when the file is copied to a partition which doesn't have NTFS (FAT/ISO9660)... Since the data is an 'attached' file, you can store any kind of data there (and access or execute them like other files/executables)
First of all, educational institutions are some of the worst places for IT people.. generally, they are run by suits and not techs. Your users are often young adults or children and professors that don't really know anything.
I'm on contract (system administration and programming) with a small internet service provider.. very organized group. Generally, clients of this ISP are very friendly and some-what knowledgable, dispite not being system admins. When I fix a problem, or even if we are unfortunate enough to not be able to fix it (corrupted backups, etc), they are very thankful and give us lots of praise.
I have been on contract and employed by other internet service providers, but none have ever had the customer satisfaction and praise that this one gets. The cause of this is quite simple, fast results. If someone emails you, reply within 10 minutes.. even if that reply is just to inform them that you're examining the problem.
Telephone support is a bad idea, especially if your user's native languages don't match your native language. Many users will refuse to stop talking about unrelated topics, others will take 10 minutes to try and tell you their password. Telephone contact for contracts, especially for programming projects isn't bad.. but it is very difficult for many kinds of technical support.
Some of my contracts are scheduled or 'on my own time', others are 'on call'. Most ISPs don't require 'on call' unless you have some special skill that the others don't have, or if they are under-staffed.
Users can't blame you for not notifying them if you're organized enough. Make a webpage with notifications and then send emails to the users notifying them of the downtime. If they complain, point them to the webpage..
I do understand the 'changes in routine' example you gave, but this is again part of organization. DO NOT make user-interface changes in 'stable' software. Prepare a new 'major' release and then make notifications of a 'major' upgrade.. and then have a meeting discussing the changes. Of course, this may seem silly if it is only a small change... but the users are less likely to complain about an 'upgrade' than 'jim playing with our software'. Users complain less about upgrades than they complain about random, unannounced changes in their configuraiton... especially if the upgrade has been properly 'hyped' (Users pay to 'upgrade' to WindowsXP from Windows98.. they are not only willing to see change, but also pay for it!)
It can be a very rewarding occupation. If you're not rewarded, then you're either in the wrong field or the wrong company.
The problem is that there is no company with a 'cheap ink, expensive printer' business model, unless you're looking for Dot-matrix.
It is questionable if forcing the companies to change this business model would be legal, ethical, etc.. However, many consumers would be very happy with this change.
Slashdot Mirror
Wow.. Debian *STABLE* releases contain older, more stable software.
So what if Debian's stable releases are long and far between? They still release much more often than Microsoft, Apple, and Sun do and with many more applications.
If you've ever ran Debian 'testing' or 'unstable', you would realize that NO other distribution or operating system is nearly as upto-date as Debian. It is only the 'stable' branch that is older, with good reason.
The GPL makes it impossible to link to non-GPL code. However, it is not illegal to distribute code that cannot be compiled.
The developers of MPlayer did not break the GPL by distributing source code. Users whom linked GPL code with non-GPL code broke the GPL.
Although it may not be 'ok' for users to make illegal binaries, the developers of the software are not to blame. The MPlayer developers simply made a note stating that it could not be compiled legally, which was responsible.
Because the Net2Phone software has the algorithm implimented in software. It is a patented algorithm and cannot be used in free software, hence free software must currently use a hardware-based solution.
The hardware apparently has some advantages over software-only implimentations, including the usage of a real telephone.
vonage.com will work for those without Microsoft Windows as it is a hardware-only solution, it connects to your hub/switch and 'just works'. The nice thing you get with GnomeMeeting is the ability to dial from the computer and the option to use the computer to speak or process the audio data.
The advantage with Quicknet vs Vontage.com is that Quicknet does not (currently?) have a monthly access fee. Vontage will charge you a monthly fee, for unlimited calling (or less for 500 domestic minutes).. Quicknet charges per-minute and is blind to national borders.
The claim that it is the first solution for linux may be true; however, it may not be true that this works in *BSD.
This requires a PCI card to work. There is an open source driver, but I don't know if it it has been ported to *BSD.
This isn't to say that it will not work in BSD soon, but it has been tested and is known to work currently, without modificaition on Linux.
Sure, knowing and distributing ips might seem like a good idea.. but it would be a nightmare for usability. You may say, "but just put the IP in the address bar of a webbrowser"; however, if that is your argument than you must have forgotten about HTTP/1.1. HTTP/1.1, and other protocols such as SMTP allow multiple domain-names exist at the same IP address. If you do not pass the domainname AND the IP with these protocols, the servers will default to god-knows-what.
We would have been ok without dns, but currently available services would be difficult or impossible to replace. Rather than typing http://192.168.1.2/, you would have to type http://somedomain.com!192.168.1.2/ or other such abdomination (@ is not used as it is popularly used to designate HTTP (and sometimes FTP) authentication).
If we didn't use DNS and such hostname-based hacks didn't appear in HTTP and SMTP, we would have a much greater need for IPV6!
DNS is a good thing, dns servers must be repaired.
Just so you're aware, one firewire bus can only do 400Mbps. You might want more than 100Mbps (~12MB/s) per drive. :)
:)
Personally, I'd go with buying an empty drive chassis with 5.25" drive bays.. then create a firewire backplane and use those ide drive caddies for holding the disks.
The only question is how many firewire controllers do you wish to have ?
Sure, it is in a VM.. but a pretty fast vm... not to mention that with the right machine you can run MacOS in a VM faster than any apple machine can.
At the cost of a little overhead, you can run MacOS under the MacOnLinux VM. Any PowerPC machine that can run linux can run MacOS 7-10.
If you don't think the fastest apple-branded machine is suitable.. how about a quad-cpu Power4 machine from IBM? They do have some powerful workstations, although the servers could be fine too as you could export the MacOnLinux display via X11 or it's (optional) internal VNC server. It supports the Linux framebuffer too if you don't want the overhead of an X11 server.
> This may work well for a school, but everybody
> around here realizes that monitors almost always
> out-live just about every other component. The iBook
Actually, I wouldn't buy an (i|e)Mac because I fear that the monitor would die leaving me with an unusable system!!
In my experience, monitors die before anything else. I went through 4 monitors in 2002... and 2 monitors so-far in 2003 (noting that it is still January!).. granted, 2/6 of those monitors were quite old (9 years)..
I've been finding that newer CRTs are less reliable than the older models. Is it sane that I have an 8month old monitor die within the same week as a 9 year old monitor? Something is obviously wrong with the manufacturing of that 8 month old monitor! Not to mention that the 8 month old monitor was a *replacement* for a monitor that was bought one year before it's death.
btw, environmental issues have been ruled out as the cause of death for these monitors.
- You do not need to buy apple to run MacOS.
- There ARE PPC motherboards, abeit expensive... but that is because there isn't a big market for them. There is a board designed for Amiga OS4 that works fine with Linux as well as the IBM machines. The importance is that any PPC board that runs linux can run MacOS, quickly too.
- OSX will not come to x86, are you nuts? Apple would not dare lose their superior architecture and try to fight Microsoft in a 1:1 battle. The best that happens from x86 OSX is that a few Windows users will dual-boot or use VMware. Apple doesn't make their money on software.
You are right about one thing, the Xserve is underpowered and over-expensive.
Except you don't find $1/GB on SCSI.. especially not SCA if you plan to hot-swap. I suspect it would be much cheaper to obtain Firewire->IDE adapters which would allow you to hot-swap.
Of course, Firewire1 isn't fast enough.. you would have to go with Firewire2 for any reasonable ammount of speed. 400Mbps
120GB * 9 = 1080GB (120GB drives are cheapest)
$170 * 9 = $1530 ($170 for a 120GB drive and firewire2 enclosure)
Each Firewire2 bus can do 800Mbps, which is 100MBps.. you would want each drive doing at least 50Mbps, requiring 5 Firewire2 controllers (and associated costs) in your machine... assuming you can FIND Firewire2 controllers (I haven't been able to, other than what is builtin to the new Apple machines). You could go with slower speeds for less money, your choice.
Of course, as the original poster said.. you can just use IDE.
Imagine how many cows/chickens/turkeys/pigs/etc we wouldn't kill if we were happy just drinking their blood... sure, we would probably have to have a lot more of those animals in captivity at one time.. but the total number of animals required for any extended period of time would be much lower..
Humans need to evolve to requiring only blood and other disposible parts.. better to hurt animals than kill them. Better than vegatables too.. vegatarians think they're so much better because they don't kill animals... they never stop to think about the plants! Plants have feelings too.
Trust me, it isn't that simple.
One would initially think that 'just blocking them' would be simple.. but it isn't. You can't block them, not within reason..
Due to NDA, I can't say more.
It is genius.
Mozilla may not have the greatest share of the market, but it may be the best browser available. This is why Apple DOES NOT want Mozilla. Sounds crazy? Not really.
Jobs realizes that competition will create better software. It would certainly be possible for Mozilla to become so popular and 'standardized' on the Unix and MacOS operating systems that development of KHTML would slow down and eventually die. If you have a company behind KHTML like Apple while AOL is behind Mozilla, you can expect a war to brew.
Mozilla is a great browser, KHTML is not bad.. but unless they become more popular and gain more press, Microsoft won't bother to compete.. they won't have to.
If KHTML and Mozilla begin a new browser war, first.. new OSX users will be using KHTML, Linux/Unix geeks will be using either Mozilla or KHTML. Apple still does have a large userbase, using KHTML could really put a dent in Microsoft. KHTML's competition would make Mozilla better and more popular, even in Microsoft Windows.
Apple may have just sparked not only a browser war, but a rejuvination of computing without Microsoft. I won't be surprised to see 30-40% of the web using non-IE browsers within a year.
Some people like the look of the TiBooks, some WANT the small form-factor, others just hate the cheaply made keyboards on the iBooks.
The iBooks, although a bit nicer looking since the clamshell design, feel like toys made of cheap plastic.
I'd rather have my G3 Powerbook with firewire (pismo) than a new iBook.
Using font-size: small should not post a significant problem with any browser if implimented properly. It is true that it may provide a smaller font than the user has requested of their browser; however, the browser should render it at a slightly smaller, yet readable size.
With the default settings of every browser I've tried (Opera5/6-mac, IE4.5/5.1-mac, Galeon/Mozilla-linux, Mozilla-mac, Netscape4-mac), font-size: small provides readable fonts..
Not that I terribly condone the practice of making the font smaller than the user's preferences.. I personally think that everyone should have their browser properly configured and let designers make their pages accessible; however, if you must make the font smaller.. font-size: small is the safest and most accessible-friendly way to change it.
YMMV.
sometimes companies use lawyers in addition to their loyers.
Realizing the 'open source' (sort of) nature of Darwin, I've thought of doing this myself (for linux).. but just didn't have them time.
For those who think this is a dumb idea.. you're not seeing the bigger picture. Currently, MacOS X only runs on G3 and G4 apple machines designated by Apple unless you boot Linux and install Mac On Linux (similiar to Plex86 for PPC). If you can run OSX applications on Linux, NetBSD, etc.. then you wouldn't need the overhead of a VM. Your RS6000 could run OSX with minimum overhead. From NetBSD you could run from the same machine, binaries compiled for: OSX, Linux, FreeBSD, and NetBSD.
Personally, I don't like OSX; However, I wouldn't mind being able to run natively on Linux: Internet Explorer, Quarx Express, etc.. Not to mention that Quartz is a pretty neat looking, X11 compatable (with Xfree86-rootless) windowing system.
This is more important for Linux than NetBSD, only because NetBSD currently has poor support for devices on many PowerPC machines (compared to Linux), especially laptops.
I write my pages with valid XHTML and CSS2. The pages will render perfectly in Mozilla and with slight tweaks, Konqueror and Opera. IE4.5 and IE5 will generally render quite well; however, IE6 is (almost?) IMPOSSIBLE to get to work.. the browser is just plain broken. It still doesn't properly render my pages.
Microsoft actually had better CSS support in IE5.x than it does in 6.
Blender. Moonlight Creator. Softimage. Lightwave. Maya Alias/Wavefront.
BTW: google.com is your friend.
> "I've also been thinking that a filesystem such as
> NTFS or ReiserFS that allows attaching huge
> ammounts of metadata, or small amounts of metadata
> to any file would be important. For instance, my
> 4096x2048 digital photograph of the grand canyon
> (big file), should still be able to have a
> thumbnail (say about 128 KB) attached as metadata."
Uh, you mean like ADS in NTFS? Some thumbnail programs use it for keeping the thumbnails attached. The data is automatically discarded by NT when the file is copied to a partition which doesn't have NTFS (FAT/ISO9660)... Since the data is an 'attached' file, you can store any kind of data there (and access or execute them like other files/executables)
Google: NTFS ADS
First of all, educational institutions are some of the worst places for IT people.. generally, they are run by suits and not techs. Your users are often young adults or children and professors that don't really know anything.
I'm on contract (system administration and programming) with a small internet service provider.. very organized group. Generally, clients of this ISP are very friendly and some-what knowledgable, dispite not being system admins. When I fix a problem, or even if we are unfortunate enough to not be able to fix it (corrupted backups, etc), they are very thankful and give us lots of praise.
I have been on contract and employed by other internet service providers, but none have ever had the customer satisfaction and praise that this one gets. The cause of this is quite simple, fast results. If someone emails you, reply within 10 minutes.. even if that reply is just to inform them that you're examining the problem.
Telephone support is a bad idea, especially if your user's native languages don't match your native language. Many users will refuse to stop talking about unrelated topics, others will take 10 minutes to try and tell you their password. Telephone contact for contracts, especially for programming projects isn't bad.. but it is very difficult for many kinds of technical support.
Some of my contracts are scheduled or 'on my own time', others are 'on call'. Most ISPs don't require 'on call' unless you have some special skill that the others don't have, or if they are under-staffed.
Users can't blame you for not notifying them if you're organized enough. Make a webpage with notifications and then send emails to the users notifying them of the downtime. If they complain, point them to the webpage..
I do understand the 'changes in routine' example you gave, but this is again part of organization. DO NOT make user-interface changes in 'stable' software. Prepare a new 'major' release and then make notifications of a 'major' upgrade.. and then have a meeting discussing the changes. Of course, this may seem silly if it is only a small change... but the users are less likely to complain about an 'upgrade' than 'jim playing with our software'. Users complain less about upgrades than they complain about random, unannounced changes in their configuraiton... especially if the upgrade has been properly 'hyped' (Users pay to 'upgrade' to WindowsXP from Windows98.. they are not only willing to see change, but also pay for it!)
It can be a very rewarding occupation. If you're not rewarded, then you're either in the wrong field or the wrong company.
Uh.. you can still use VIM
The problem is that there is no company with a 'cheap ink, expensive printer' business model, unless you're looking for Dot-matrix.
It is questionable if forcing the companies to change this business model would be legal, ethical, etc.. However, many consumers would be very happy with this change.
--
Eric Windisch
I think tickets in my area (the Philadelphia 'burbs) are over $7 now.. certainly not $3.25, even for the elderly students during matinee hours. :)