firstly can I just say "gosh, a lucid and worth while reply to one of my lengthy posts - thanks, made my day * smile *", thanks!
I think we're both saying the same thing but from slightly different angles, perhaps you more capably than I in fact;-)
But I'd like to just touch on a couple of the points again if I could.
Just because it's "used" as a noun, doesn't in turn make it right, I refer to varous forms of Traffic Authorities as a range of four letter "nouns" but that doesn't make the Police department become what I call them.
As a descriptivist, I have to disagree with you here. If enough people use a word to mean something, then that's a perfectly valid meaning of the word. If you look up "pig" in most dictionaries, you'll see a definition there which says something like "Offensive Slang. Used as a disparaging term for a police officer." Enough people use it that way, so it's a perfectly valid term. Likewise, if you happen to have a dictionary which contains "Linux" in it, most likely, if it wasn't written by the FSF, it will say that Linux is an operating system, not a kernel, because that's what people use the term to mean. In fact, that's the what the person who originally invented the term meant, and it's what the etymology of the term shows that it means. You don't dispute that "Minix" is an operating system, and not just a kernel, do you?
You're right, and I concur on most of your responce, in that agreed, common use will force a term or name into day to day language, and that's more likely to increase as products and brands continue to push into common language through heafty investment in merging into our day to day lives.
Nowhere more so than in technology, in fact in technology due to the fact that people don't on the whole understand what they are talking about ( non techs that is, like my mum, or father in law in particular who both love what I do for a living and try to say in touch with it but really don't understand what I do - then again neither do I some days * grin * ), so they clasp onto any term they can and more so if it's a "handle" or name per se.
As for Minix, yes I do consider Minix a whole OS as such, partucularly in light of the fact that AT wrote it all from scratch ( well with some student help I think ) and the kernel and the OS tools etc including the file system were all part of the whole, that is, he didn't take seperate kernel, file system, os tools and drivers, and unix like commands and glue them together to make Minix.
Minix was built from ground up at source level, and like say FreeBSD, or the other BSD's, BeOS, NeXT Step, and others, they are in their own right "whole" operating systems. So I don't have an issue calling Minix an operating system and I love that the Minix kernel is part of the whole.
Solaris even I consider a whole operating system, albeit the fact that it's roots were in the BSD family, but through market forces it was driven into the SYSV camp, I really do consider Solaris with it's tighly controlled kernel and os software, drivers, libraries, and unix tools, to be a whole and complete operating system.
Mac OS X on the other hand, is bordering on what I see as a complete OS that I would, like GNU/Linux otherwise refer to in it's parts.
Mac OS X 10.x.x for example, with it's Mach micro kernel, it's BSD 4.4 core, NeXT Step inherited file system and library layout, GUI hand me downs, glued together to form the OS, with Mach and Darwin glued as much as they can be, and the GUI layers stacked high and deep, it is a jumble, but at the end of the day it's glued together pretty well now and the gap between the kernel, the file system, and the OS software and tools is pretty narrow and clean.
But I for one tend to use the Brand name of the "distro" rather than just refer to a distro as Linux, for example
And what if you roll your own? What should I call my Linux From Scratch distro? I
Just because it's "used" as a noun, doesn't in turn make it right, I refer to varous forms of Traffic Authorities as a range of four letter "nouns" but that doesn't make the Police department become what I call them.
I think we'll find that marketing departments at Red Hat and their like, and the media at large, in want of a short name ( to be catchy, pithy, and save column space ) also pushed the use of Linux to generally refer to the whole as Linux.
Look, I like all of you, use Linux to refer to the GNU/Linux "operating system", but it's never felt right - no matter how much I use it.
Solaris, FreeBSD, NetBSD, and OpenBSD even, Mac OS X, and god forbit Winows for some reason, all feer right, but Linux never has, and it's bugged me since the day I first downloaded an almost stable kernel to run up.
I think we're stuck with Linux as a name, I'll grant you that, in the same way that we're stuck with Google being the media driven, and I'm sure Google marketing department, and most likely desperate for a word to describe what they are doing ludites ( that's not really the right name for them either I guess ) who wanted a single word/noun to describe their web searching ( why web search didn't stick I don't know - perhaps we all have indeed been brain washed by productization to want a brand name to tag onto rather than be satisfied that the english language will more than suffice on it's own merit? ).
But I for one tend to use the Brand name of the "distro" rather than just refer to a distro as Linux, for example:
Red Hat SuSE Mandrake
et al
I feel more comfortable and correct when I say that I run Red Hat, or that I run say SuSE, as apposed to saying I run "Linux" per se, as the very next question I get when I do say I run Linux, is "which distro?".
See?
So perhaps rather than GNU/Linux, we should be saying that we run a Distro Name, such as Mandrake, when we refer to Linux distro's, and just drop the whole Linux bit?
Let's face it, there's great distro's like Debian for example that are easy to run up using other kernels than Linux.
quote from www.debian.org:
" What is Debian?
Debian is a free operating system (OS) for your computer. An operating system is the set of basic programs and utilities that make your computer run. Debian uses the Linux kernel (the core of an operating system), but most of the basic OS tools come from the GNU project; hence the name GNU/Linux. "
So if Debian is up front and clear about their OS / Distro running a Linux "kernel" but that "most of the basic OS tools come from the GNU project", "hence the name GNU/Linux" then gosh guys, even the distro folk are telling us that it's GNU/Linux!
But before you miss the point of the last bit there, Debian as an operating system does for example run just fine with other "kernels" than Linux!
quote from the www.debian.org/ports/hurd/ page:
" Debian GNU/Hurd
The Hurd is a set of servers running on top of the GNU Mach microkernel. Together they build the base for the GNU operating system.
Currently, Debian is only available for Linux, but with Debian GNU/Hurd we have started to offer GNU/Hurd as a development, server and desktop platform, too. However, Debian GNU/Hurd is not officially released yet, and won't be for some time "
Then there's the gem that really puts all this into perspective folks, and that's from the Debian page:
quote from www.debian.org/ports/
" Introduction
As most of you know, Linux is just a kernel. And, for a long time, the Linux kernel ran only on the Intel x86 series of machines, from the 386 up (there is work being done to port Linux to 286, and earlier, machines. See the ELKS project for more information).
However, this is no longer true, by any means. The Linux kernel has now been ported to a large, and growing, list of architectures. Following close behind, we have ported the Debian distribution to these architectures. In general, this
I must have missed something, or was it just mass media brain washing that has caught on? But last time I looked, when I installed something like SuSE, Red Hat, or Debian, it was an operating system built on open source tools, which compirsed of the linux "kernel", some variant of the unix file system, a whole suit of gnu replacements for unix commands, and a range of open source packages from folk like Apache and such?
If we were to talk about perhaps Solaris, then indeed, we are talking about the Solaris kernel, the Solaris operating system tools which were all written from scratch, alebit with access to the source from BSD and SYS V variants, and agian a unix file system and some packages from folk like Apache and such, but in this case it's a complete solution from Sun and it's called Solaris.
The same can be said surely for the likes of OpenBSD, NetBSD, and FreeBSD, where they are complete systems, built around kernels, from scratch, although in each case they too lean heavity on the GNU replacements for Unix commands and tools.
Windows for example once refered to itself as Windows NT, where the NT part was essentially the kernel, designed and built by some smart folk who had a hand in the likes of OS/2 and VMS kernels and operating systems if I recall corrently, but it was clear that Windows was the GUI and NT was the underlying kernel.
Mac OS X even now is pretty open about the split between it's Mach kernel, Darwin core, and BSD / NeXT Step tools, but we don't call Mac OS X "Mach" do we - nope, it's OS X, or if you're like me and you favour what uname -a tells you, it's Darwin;-)
I think Stallman summed it up pretty well when he ended the piece with:
quote:
Stallman thinks the issue of naming the product is not so clear cut. "Most of the time, when people call something 'Linux', it's the GNU system with Linux as the kernel. Maybe this policy will encourage people to call it GNU," Stallman told the Sydney Morning Herald. "I prefer to say GNU/Linux' so as to give the kernel's developer a share of the credit."
Now I do agree that GNU/Linux is perhaps a mouthfull, but on the other hand, I think it's particularly lame to refer to the GNU/Linux operating system as just Linux, so perhaps it's time for a new name, label, whatever, for whatever it is many of us run.
It could be like the Musician formerly known as Prince, now known as some Egyptian hyrogliph - we could have the operating system formerly known as Linux, now known as #$%^&#!?
It might actually be worth many of you taking time to read Stallman's FAQ on GNU/Linux over at:
It does go a long way to answering and clearing up much of what is in this horribly messy series of threads and sub threads, basically emotive and guess work, rather than fact.
For example, from that URL:
quote:
Why do you call it GNU/Linux and not Linux? Most operating system distributions based on Linux as kernel are basically modified versions of the GNU operating system. We began developing GNU in 1984, years before Linus Torvalds started to write his kernel, and we developed a larger part of the resulting system than any other project. In fairness, we ought to get equal mention.
quote:
Why is the name important? Although the developers of Linux, the kernel, are contributing to the free software community, many of them do not care about freedom. People who think the whole system is Linux tend to get confused and assign to those developers a role in the history of our community which they did not actually play. Then they give inordinate weight to those developers' views. Calling the system GNU/Linux recognizes the role that our idealism played in building our community, and helps the public recognize the practical importance of these ideals.
when they say "companies" they are only refering to companies "within" the Northern American continent.
I think it's fair to say that outside of Northern America, comapanies as it were are hardly likely to "outsource" to service providers in "Small Town USA" as apposed to the likes of India, Indonesia, Ireland, Wales, China, et al.
I could be horribly wrong, but I'd say that on average the level of education of Indian's working in India based call centers for example would be higher than that of rural USA - I hope I'm wrong for the USA's sake but I suspect I'm not ( time to do the home work I guess Dez ).
But most Indian's know where other countries are on the map for example, where as most Americans, rural USA that is, don't know many countries outside of the USA, so I'd rather not send my follow the sun support to rural USA if the local's are not able to even picture in their mind where my Australian customers are on the map;-(
Weighing in with my two cents worth, for what it's worth, I'd like to brain dump what I would consider worth while options for your needs. All of these are solutions I either have used in the past successfully, or am currently using for various purposes. So bear in mind that this is not just the causal musings of a thread cruiser, but actual tried and proven solutions;-)
First some basic assumptions:
1) You want to run some form of Unix or Unix like system ( i.e. Linux ) - you've noted you currently use your Apple PowerBook laptop, so one has to assume you're running Mac OS X 10.x.x natively ( more power to you ).
2) You want complete control over the system including "root" access 24/7 - this is of course the whole point of having your own system, you can beat it up, break it, rebuild it, and all that jazz.
3) The system should be able to be run remotely, even if just headless on your LAN, or perhaps more ideally remotely from some external 3rd party in a hosted solution so you don't end up having to host it behind your link at home ( also making it easier for you to provide access to other parties should you want to either share it with friends and family or if you just want to make it world visible for whatever reason - i.e. your own mail and web server et al ).
4) You want an "always on" solution, so this should be something that, as you state, should not suck too much juice power wise, is able to be built with a "standard build" style hardened platform, which in the case of power loss would ideally recover nicely, quickly, and be back on line ( I'll touch on this later as standard builds are going to make your life so much simpler and fun ).
5) The performance of the system ideally should be such that it will cope with the key elements you've noted in your post, such as:
a) remote access such as remote sessions via SSH won't kill the system
Ok, so Russian cosmonaut, Sergei Krikalev has taken the record for most time spent in space with a whopping 748 days in space.
Now that's really quite something when you consider the technology that we're putting folk up into space with, particularly in light of recent issues getting back with Shuttle missions.
Indeed his ability to deal with the physical hardships in space is quite something, when you consider that under a single atmospheric preassure of 1G, the average American is unable to remain in shape (and tending towards obese), Australian's are unfortunately not much better frankly.
So I can only imagine what the strain of keeping up with the challenge of what must be a massively disciplined exercise regime.
Check out the whole story over at New Scientist magazine online.
My ongoing consern though, and this record simply brings it back to mind, is that if we've got the world's best out there in near earth orbit, pulling just 748 days so far non-stop.
And that's without the rigors of deep space travel, heck, that's without what should be a reasonably simple challenge of travel within the local group of planets!
How exactly are we going to eventually break the bounds of our planet and travel out to the local planets, and beyond?
Space is big, Really big. And that's just the local group, let's not boggle our minds with the really really really big, let's just stick with the really really big.
Interplanetary travel involves distances of hundreds and thousands of millions of kilometers between fast moving objects (planets for example).
For example, the Earth travels around our sun at an average sleep around 30 km/sec, Mars is traveling at around 24 km/sec (so that's a difference of about 20,000 kilometers per hour!), and the distance between them is millions of kilometers. Phew - brain hurting yet?
The legal definition of space is 50 kilometers up, but for our purposes, "space" starts about 200 kilometers up.
There's still a heck of a lot of atmosphere ("lot" being a relative term) there, but an average satellite can stay in orbit for weeks before the atmospheric friction slows it enough so it falls out of orbit.
I don't want to belittle what is in todays terms simply a herculean effort. 748 days is 748 days, simple as that, and it's the current record, and more power to Sergei Krikalev frankly.
But at that rate, and we're talking a mere couple of years really (2.04 years), that's not even a one way trip to some of our nearest neighbors, let alone the return trip, if you were thinking of coming back that is * grin *
We are though using developments from space technology to do some pretty weird things, for example, the industrialization of a NASA-tested concept for artificially creating meat!
According to a recently published academic paper, Edible Meat Can be Grown in a Lab on Industrial Scale. ( read more about it at GizMag ).
I wonder if vegies and vegans will now consider eating meat if it doesn't involve killing a living breathing animal? Doubt it.
Anyway - Three cheers for Sergei, but as for the rest of the space program, no dice, we're clearly generations away from truely being a space faring people.
It's been a long time coming, but finally it looks like Telstra will be split in two very seperate companies.
One to provide Wholesale Telecoms and the other to provide Retail Telecoms services.
This is so horribly over due!
We may be on the verge, finally, of seeing some real competition in the Australian teleco market place.
Telstra naturally are not very happy about what is now an approved [by cabinet] package, which will force them [Telstra] to create seperage network and retail divisions, with separate premises and management but under the same company structure.
If all goes to plan, the government (coalition) could steamroll ahead and sell its majority share holding ( 51.8 per cent stake ) in Telstra before the end of next year. Something which has been high on the governments agenda for some time now.
Sure that's going to make them a truck load of money, it's (the sale of 51.8% of Telstra is a lot of share value) going to do that no matter how you look at it.
What does all this mean for Australia?
Well we could perhaps look just accross the Tasman to the windy city of Wellington, in New Zealand, for an example of just what a completely deregulated teleco marketplace can do if you allow it.
Businesses and individual end users in Wellington, can gain access to Data and Voice services that the rest of the world ( except perhaps for Singapore with their Interent Corridor ) dreams of.
With 10 megabit and 100 megabit, and even gigabit connections for tens or hundreds of dollars a month, zero data usage charges, and peering for one and all if you want it, Wellington has shown that a completely deregulated telecommunications industry can work, and will work, if you allow it to.
We won't in the near future, say the next five years even, see the likes of what NZ has been able to achieve here in Australia, well not from what I can see gazing into my crystal ball anyway, as there's a legacy culture to be left behind before we can see Australia make major leaps forward.
I'm hoping that with Telstra now having to form a legitimate Wholesale arm, freed up and allowed to sell outside of it's previous one and only customer, being Telstra itself ( oh and the occasional carrier and ISP when they had time of course ), Telstra Wholesale may be allowed to sell core services at prices that would allow 3rd party providers, in particular the DSL providers, or the Broadband market at least, provide now ADSL 2+ services of 22 megabit speeds, throughout the country at prices equal to what we now pay for 1.5 megabit links.
What does the general media have to report? Here's a few links for further homework on the topic:
Let's hope that with new management, and a sense of responsibility to the nation, the new Telstra's can both give back a little of what they have so easily come by, and finally deliver on the government's Digital Nation promise.
I'm going to clock in here with my experience to date with m0n0wall which has been fantastic ( no I don't own shares in anything to do with m0n0wall *grin* - wish I did !! ).
I have to say that from my experience to date with it, m0n0wall is without a doubt one of, if not THE, leading firewall platforms currently available in the open source world, and it's fair to say that I've had a thing or two to do with firewalls and security in general over the past 20+ odd years.
with years and years of hands on design and implementation using checkpoint on sun, checkpoint on nokia, cisco routers, cisco pix, netscreen, ipf, ipfw, iptables, blah blah.
heck, I had such a hard on for checkpoint that at one stage I've even run up a SOFAware box which has the checkpoint inspection module in it, although it's web interface is crap and you can't actually do anything with the firewall policy other than port mapping and translations.
anyway the bastard thing kept resetting and or just slowing down to the point of being so useless I threw it away - after putting it through a hammer test - hammer won * grin *
so I've played with firewalls ok, and god knows how many other bloody firewall platforms, I've played with as many open source firewalls as I can get my hands on, and m0n0wall in particular really has impressed me. When I say play by the way, I mean I've put it through some horrible lab testing, really pushed till smoke came out of the things!
note: firewall blog with reviews of the various firewalls pending kids;-)
smoothwall in my experience had made some very serious inroads towards what was going to become a very strong contender, but then the group fell into ( from what I could tell from the sidelines ) a political infighting jihad which still effects the project.
add to this that they [in my opinion] seemed to have also very seriously stuffed up with their DSL support in 2.x by only supporting USB models of the more widely used DSL modems, particularly here in Australia where Alcatel Speedtouch modems are used far and wide.
in fact it was during an upgrade attempt from smoothwall 1.x to 2.x, I found this out when I was trying to get my DSL modem to talk to smoothwall etc, and out of sheer frustration I decided it was time to dump smoothwall and have another look around.
for a time I even tried running iptables on linux, using fwbuilder on my mac natively and seriously hardened redhat 7.3 ( lord knows it needed it ), horribly stripped down with just enough of the base os left to support two ethernet cards, iptables, and ssh ( to allow fwbuilder to install it's policy ), and I'm still a very big fan of this model, but the one thing that I found a headache setting up and maintaining using fwbulder in this sort of architecture was vpn connections / clients. Also shaping traffic wasn't really feasible and nobody in their right might these days ( again my personal opinion ) runs anything on a network without some form of shaping! Do they?
so again I went hunting the open source tundra for a new toolset. this was when I re-discovered m0n0wall, which when I first reviewed it, was perhaps at a very early stage in it's life cycle and by no means the magical wonderland that it is todya [as of 1/6/2005 (that's July 1st for you American date centric folk)].
Key strengths that I've had working and under high loads, include:
- base firewall policy made up of some very complex rules - multiple dmz's ( I hate dmz's - they are lame but so be it ) - nat on wan interface, and one of the dmz interfaces - multiple static routes - multiple dynamic routes - dynamic dns ( had to tinker to get no-ip.com working but hey ) - dns caching / forwarding - ipsec and pptp vpn connections with many vpn clinets - traffic shaping with QoS which actually works! yea, it really does! - address aliases on floating ip's for fail over / redundancy - dhcp with pool of ip's as well as fixed MAC map's and static ip's - proxy
I'm going to clock in here with my experience to date with m0n0wall which has been fantastic ( no I don't own shares in anything to do with m0n0wall *grin* - wish I did !! ).
I have to say that from my experience to date with it, m0n0wall is without a doubt one of, if not THE, leading firewall platforms currently available in the open source world, and it's fair to say that I've had a thing or two to do with firewalls and security in general over the past 20+ odd years.
with years and years of hands on design and implementation using checkpoint on sun, checkpoint on nokia, cisco routers, cisco pix, netscreen, ipf, ipfw, iptables, blah blah.
heck, I had such a hard on for checkpoint that at one stage I've even run up a SOFAware box which has the checkpoint inspection module in it, although it's web interface is crap and you can't actually do anything with the firewall policy other than port mapping and translations.
anyway the bastard thing kept resetting and or just slowing down to the point of being so useless I threw it away - after putting it through a hammer test - hammer won * grin *
so I've played with firewalls ok, and god knows how many other bloody firewall platforms, I've played with as many open source firewalls as I can get my hands on, and m0n0wall in particular really has impressed me. When I say play by the way, I mean I've put it through some horrible lab testing, really pushed till smoke came out of the things!
note: firewall blog with reviews of the various firewalls pending kids;-)
smoothwall in my experience had made some very serious inroads towards what was going to become a very strong contender, but then the group fell into ( from what I could tell from the sidelines ) a political infighting jihad which still effects the project.
add to this that they [in my opinion] seemed to have also very seriously stuffed up with their DSL support in 2.x by only supporting USB models of the more widely used DSL modems, particularly here in Australia where Alcatel Speedtouch modems are used far and wide.
in fact it was during an upgrade attempt from smoothwall 1.x to 2.x, I found this out when I was trying to get my DSL modem to talk to smoothwall etc, and out of sheer frustration I decided it was time to dump smoothwall and have another look around.
for a time I even tried running iptables on linux, using fwbuilder on my mac natively and seriously hardened redhat 7.3 ( lord knows it needed it ), horribly stripped down with just enough of the base os left to support two ethernet cards, iptables, and ssh ( to allow fwbuilder to install it's policy ), and I'm still a very big fan of this model, but the one thing that I found a headache setting up and maintaining using fwbulder in this sort of architecture was vpn connections / clients. Also shaping traffic wasn't really feasible and nobody in their right might these days ( again my personal opinion ) runs anything on a network without some form of shaping! Do they?
so again I went hunting the open source tundra for a new toolset. this was when I re-discovered m0n0wall, which when I first reviewed it, was perhaps at a very early stage in it's life cycle and by no means the magical wonderland that it is todya [as of 1/6/2005 (that's July 1st for you American date centric folk)].
Key strengths that I've had working and under high loads, include:
- base firewall policy made up of some very complex rules - multiple dmz's ( I hate dmz's - they are lame but so be it ) - nat on wan interface, and one of the dmz interfaces - multiple static routes - multiple dynamic routes - dynamic dns ( had to tinker to get no-ip.com working but hey ) - dns caching / forwarding - ipsec and pptp vpn connections with many vpn clinets - traffic shaping with QoS which actually works! yea, it really does! - address aliases on floating ip's for fail over / redundancy - dhcp with pool of ip's as well as fixed MAC map's and static ip's - proxy
I think those of us with Mac OS X and half a klew already use standards compliant methods to make sure scripts work on any platform, i.e.:
#!/bin/sh - echo "Hello World!" exit 0
99.2% of people using Mac OS X prolly don't use a shell anyway.
With 18+ years unix I've gone from hard copy tty line printer consoles, to vt100's, wyse 75, wysel 180, to a cranky old toshiba 5200 plazma screen ( loved that ) running Xenix, then SCO, then 386/ix, and eventually BSD/386. I tried using a toshiba techra 8200 with windows 95, 98, 2000 and it was crap, then went to using an old Mac Powerbook 150 and never looked back - I've always had a mac around the place, but once I started using a mac, I found it the closest thing to having a sun workstation in a laptop, without the cli ( bugger ) - I even ran Apple's "Apple UX" BSD system once, but it was heavy on old mac's, even ran the tenon stuff - same deal.
So when Mac OS X 10.0 came out, I dropped everything and ran to the nearest ( yes, I said ran! ) shop, and got a 15 inch G4 and have never looked back. Now I can't understand why people run PC platform latop's with Linux when they can have the perfect world of a Mach kernel ( the best ), a FreeBSD core with the whole of the open source world at their feet, and everything I have ever needed that ran on Linux runs on Darwin just fine ( unless some moron has written pissy code and then they can get stuffed - I'll write it myself ).
Convert now - dump intel, bring your linux thinking, and run Darwin on PPC and see a brighter world!
Slashdot Mirror
Hi Anthony,
firstly can I just say "gosh, a lucid and worth while reply to one of my lengthy posts - thanks, made my day * smile *", thanks!
I think we're both saying the same thing but from slightly different angles, perhaps you more capably than I in fact ;-)
But I'd like to just touch on a couple of the points again if I could.
Just because it's "used" as a noun, doesn't in turn make it right, I refer to varous forms of Traffic Authorities as a range of four letter "nouns" but that doesn't make the Police department become what I call them.
As a descriptivist, I have to disagree with you here. If enough people use a word to mean something, then that's a perfectly valid meaning of the word. If you look up "pig" in most dictionaries, you'll see a definition there which says something like "Offensive Slang. Used as a disparaging term for a police officer." Enough people use it that way, so it's a perfectly valid term. Likewise, if you happen to have a dictionary which contains "Linux" in it, most likely, if it wasn't written by the FSF, it will say that Linux is an operating system, not a kernel, because that's what people use the term to mean. In fact, that's the what the person who originally invented the term meant, and it's what the etymology of the term shows that it means. You don't dispute that "Minix" is an operating system, and not just a kernel, do you?
You're right, and I concur on most of your responce, in that agreed, common use will force a term or name into day to day language, and that's more likely to increase as products and brands continue to push into common language through heafty investment in merging into our day to day lives.
Nowhere more so than in technology, in fact in technology due to the fact that people don't on the whole understand what they are talking about ( non techs that is, like my mum, or father in law in particular who both love what I do for a living and try to say in touch with it but really don't understand what I do - then again neither do I some days * grin * ), so they clasp onto any term they can and more so if it's a "handle" or name per se.
As for Minix, yes I do consider Minix a whole OS as such, partucularly in light of the fact that AT wrote it all from scratch ( well with some student help I think ) and the kernel and the OS tools etc including the file system were all part of the whole, that is, he didn't take seperate kernel, file system, os tools and drivers, and unix like commands and glue them together to make Minix.
Minix was built from ground up at source level, and like say FreeBSD, or the other BSD's, BeOS, NeXT Step, and others, they are in their own right "whole" operating systems. So I don't have an issue calling Minix an operating system and I love that the Minix kernel is part of the whole.
Solaris even I consider a whole operating system, albeit the fact that it's roots were in the BSD family, but through market forces it was driven into the SYSV camp, I really do consider Solaris with it's tighly controlled kernel and os software, drivers, libraries, and unix tools, to be a whole and complete operating system.
Mac OS X on the other hand, is bordering on what I see as a complete OS that I would, like GNU/Linux otherwise refer to in it's parts.
Mac OS X 10.x.x for example, with it's Mach micro kernel, it's BSD 4.4 core, NeXT Step inherited file system and library layout, GUI hand me downs, glued together to form the OS, with Mach and Darwin glued as much as they can be, and the GUI layers stacked high and deep, it is a jumble, but at the end of the day it's glued together pretty well now and the gap between the kernel, the file system, and the OS software and tools is pretty narrow and clean.
But I for one tend to use the Brand name of the "distro" rather than just refer to a distro as Linux, for example
And what if you roll your own? What should I call my Linux From Scratch distro? I
Just because it's "used" as a noun, doesn't in turn make it right, I refer to varous forms of Traffic Authorities as a range of four letter "nouns" but that doesn't make the Police department become what I call them.
I think we'll find that marketing departments at Red Hat and their like, and the media at large, in want of a short name ( to be catchy, pithy, and save column space ) also pushed the use of Linux to generally refer to the whole as Linux.
Look, I like all of you, use Linux to refer to the GNU/Linux "operating system", but it's never felt right - no matter how much I use it.
Solaris, FreeBSD, NetBSD, and OpenBSD even, Mac OS X, and god forbit Winows for some reason, all feer right, but Linux never has, and it's bugged me since the day I first downloaded an almost stable kernel to run up.
I think we're stuck with Linux as a name, I'll grant you that, in the same way that we're stuck with Google being the media driven, and I'm sure Google marketing department, and most likely desperate for a word to describe what they are doing ludites ( that's not really the right name for them either I guess ) who wanted a single word/noun to describe their web searching ( why web search didn't stick I don't know - perhaps we all have indeed been brain washed by productization to want a brand name to tag onto rather than be satisfied that the english language will more than suffice on it's own merit? ).
But I for one tend to use the Brand name of the "distro" rather than just refer to a distro as Linux, for example:
Red Hat
SuSE
Mandrake
et al
I feel more comfortable and correct when I say that I run Red Hat, or that I run say SuSE, as apposed to saying I run "Linux" per se, as the very next question I get when I do say I run Linux, is "which distro?".
See?
So perhaps rather than GNU/Linux, we should be saying that we run a Distro Name, such as Mandrake, when we refer to Linux distro's, and just drop the whole Linux bit?
Let's face it, there's great distro's like Debian for example that are easy to run up using other kernels than Linux.
quote from www.debian.org:
"
What is Debian?
Debian is a free operating system (OS) for your computer. An operating system is the set of basic programs and utilities that make your computer run. Debian uses the Linux kernel (the core of an operating system), but most of the basic OS tools come from the GNU project; hence the name GNU/Linux.
"
So if Debian is up front and clear about their OS / Distro running a Linux "kernel" but that "most of the basic OS tools come from the GNU project", "hence the name GNU/Linux" then gosh guys, even the distro folk are telling us that it's GNU/Linux!
But before you miss the point of the last bit there, Debian as an operating system does for example run just fine with other "kernels" than Linux!
quote from the www.debian.org/ports/hurd/ page:
"
Debian GNU/Hurd
The Hurd is a set of servers running on top of the GNU Mach microkernel. Together they build the base for the GNU operating system.
Currently, Debian is only available for Linux, but with Debian GNU/Hurd we have started to offer GNU/Hurd as a development, server and desktop platform, too. However, Debian GNU/Hurd is not officially released yet, and won't be for some time
"
Then there's the gem that really puts all this into perspective folks, and that's from the Debian page:
quote from www.debian.org/ports/
"
Introduction
As most of you know, Linux is just a kernel. And, for a long time, the Linux kernel ran only on the Intel x86 series of machines, from the 386 up (there is work being done to port Linux to 286, and earlier, machines. See the ELKS project for more information).
However, this is no longer true, by any means. The Linux kernel has now been ported to a large, and growing, list of architectures. Following close behind, we have ported the Debian distribution to these architectures. In general, this
Linux is a kernel, right?
;-)
When did Linux become the operating system?
I must have missed something, or was it just mass media brain washing that has caught on? But last time I looked, when I installed something like SuSE, Red Hat, or Debian, it was an operating system built on open source tools, which compirsed of the linux "kernel", some variant of the unix file system, a whole suit of gnu replacements for unix commands, and a range of open source packages from folk like Apache and such?
If we were to talk about perhaps Solaris, then indeed, we are talking about the Solaris kernel, the Solaris operating system tools which were all written from scratch, alebit with access to the source from BSD and SYS V variants, and agian a unix file system and some packages from folk like Apache and such, but in this case it's a complete solution from Sun and it's called Solaris.
The same can be said surely for the likes of OpenBSD, NetBSD, and FreeBSD, where they are complete systems, built around kernels, from scratch, although in each case they too lean heavity on the GNU replacements for Unix commands and tools.
Windows for example once refered to itself as Windows NT, where the NT part was essentially the kernel, designed and built by some smart folk who had a hand in the likes of OS/2 and VMS kernels and operating systems if I recall corrently, but it was clear that Windows was the GUI and NT was the underlying kernel.
Mac OS X even now is pretty open about the split between it's Mach kernel, Darwin core, and BSD / NeXT Step tools, but we don't call Mac OS X "Mach" do we - nope, it's OS X, or if you're like me and you favour what uname -a tells you, it's Darwin
I think Stallman summed it up pretty well when he ended the piece with:
quote:
Stallman thinks the issue of naming the product is not so clear cut. "Most of the time, when people call something 'Linux', it's the GNU system with Linux as the kernel. Maybe this policy will encourage people to call it GNU," Stallman told the Sydney Morning Herald. "I prefer to say GNU/Linux' so as to give the kernel's developer a share of the credit."
Now I do agree that GNU/Linux is perhaps a mouthfull, but on the other hand, I think it's particularly lame to refer to the GNU/Linux operating system as just Linux, so perhaps it's time for a new name, label, whatever, for whatever it is many of us run.
It could be like the Musician formerly known as Prince, now known as some Egyptian hyrogliph - we could have the operating system formerly known as Linux, now known as #$%^&#!?
It might actually be worth many of you taking time to read Stallman's FAQ on GNU/Linux over at:
http://www.gnu.org/gnu/gnu-linux-faq.html
It does go a long way to answering and clearing up much of what is in this horribly messy series of threads and sub threads, basically emotive and guess work, rather than fact.
For example, from that URL:
quote:
Why do you call it GNU/Linux and not Linux?
Most operating system distributions based on Linux as kernel are basically modified versions of the GNU operating system. We began developing GNU in 1984, years before Linus Torvalds started to write his kernel, and we developed a larger part of the resulting system than any other project. In fairness, we ought to get equal mention.
quote:
Why is the name important?
Although the developers of Linux, the kernel, are contributing to the free software community, many of them do not care about freedom. People who think the whole system is Linux tend to get confused and assign to those developers a role in the history of our community which they did not actually play. Then they give inordinate weight to those developers' views.
Calling the system GNU/Linux recognizes the role that our idealism played in building our community, and helps the public recognize the practical importance of these ideals.
quote:
Of course this is USA centric right?
;-(
when they say "companies" they are only refering to companies "within" the Northern American continent.
I think it's fair to say that outside of Northern America, comapanies as it were are hardly likely to "outsource" to service providers in "Small Town USA" as apposed to the likes of India, Indonesia, Ireland, Wales, China, et al.
I could be horribly wrong, but I'd say that on average the level of education of Indian's working in India based call centers for example would be higher than that of rural USA - I hope I'm wrong for the USA's sake but I suspect I'm not ( time to do the home work I guess Dez ).
But most Indian's know where other countries are on the map for example, where as most Americans, rural USA that is, don't know many countries outside of the USA, so I'd rather not send my follow the sun support to rural USA if the local's are not able to even picture in their mind where my Australian customers are on the map
Dez
Prez Bush said it best when he said..
"You can fool me once, you can fool me tw.. you can't fool me!"
(you'd think his speach writers would have gotten the "You can fool me once, you can't fool me twice" bit right wouldn't you?)
Dez
http://www.blanchfield.com.au/blog/
Ok,
;-)
.html ( written in Java )
Weighing in with my two cents worth, for what it's worth, I'd like to brain dump what I would consider worth while options for your needs. All of these are solutions I either have used in the past successfully, or am currently using for various purposes. So bear in mind that this is not just the causal musings of a thread cruiser, but actual tried and proven solutions
First some basic assumptions:
1) You want to run some form of Unix or Unix like system ( i.e. Linux ) - you've noted you currently use your Apple PowerBook laptop, so one has to assume you're running Mac OS X 10.x.x natively ( more power to you ).
2) You want complete control over the system including "root" access 24/7 - this is of course the whole point of having your own system, you can beat it up, break it, rebuild it, and all that jazz.
3) The system should be able to be run remotely, even if just headless on your LAN, or perhaps more ideally remotely from some external 3rd party in a hosted solution so you don't end up having to host it behind your link at home ( also making it easier for you to provide access to other parties should you want to either share it with friends and family or if you just want to make it world visible for whatever reason - i.e. your own mail and web server et al ).
4) You want an "always on" solution, so this should be something that, as you state, should not suck too much juice power wise, is able to be built with a "standard build" style hardened platform, which in the case of power loss would ideally recover nicely, quickly, and be back on line ( I'll touch on this later as standard builds are going to make your life so much simpler and fun ).
5) The performance of the system ideally should be such that it will cope with the key elements you've noted in your post, such as:
a) remote access such as remote sessions via SSH won't kill the system
b) able to run a web server such as:
thttpd: http://www.acme.com/software/thttpd/
Apache: http://www.apache.org/
mathopd: http://mathop.diva.nl/
Roxen: http://www.roxen.com/
Boa: http://www.boa.org/
Jigsaw: http://www.w3.org/Jigsaw/ ( written in Java )
Acme.Serve: http://www.acme.com/java/software/Acme.Serve.Serve
CERN: http://www.w3.org/hypertext/WWW/Daemon/Status.html
NCSA: http://hoohoo.ncsa.uiuc.edu/
Netscape FastTrack: http://home.netscape.com/ ( not sure if it's still available )
Netscape Enterprise: http://home.netscape.com/ ( not sure if it's still available )
Zeus: http://www.zeus.co.uk/
source: http://www.acme.com
Ok, so Russian cosmonaut, Sergei Krikalev has taken the record for most time spent in space with a whopping 748 days in space.
Now that's really quite something when you consider the technology that we're putting folk up into space with, particularly in light of recent issues getting back with Shuttle missions.
Indeed his ability to deal with the physical hardships in space is quite something, when you consider that under a single atmospheric preassure of 1G, the average American is unable to remain in shape (and tending towards obese), Australian's are unfortunately not much better frankly.
So I can only imagine what the strain of keeping up with the challenge of what must be a massively disciplined exercise regime.
Check out the whole story over at New Scientist magazine online.
My ongoing consern though, and this record simply brings it back to mind, is that if we've got the world's best out there in near earth orbit, pulling just 748 days so far non-stop.
And that's without the rigors of deep space travel, heck, that's without what should be a reasonably simple challenge of travel within the local group of planets!
How exactly are we going to eventually break the bounds of our planet and travel out to the local planets, and beyond?
Space is big, Really big. And that's just the local group, let's not boggle our minds with the really really really big, let's just stick with the really really big.
Interplanetary travel involves distances of hundreds and thousands of millions of kilometers between fast moving objects (planets for example).
For example, the Earth travels around our sun at an average sleep around 30 km/sec, Mars is traveling at around 24 km/sec (so that's a difference of about 20,000 kilometers per hour!), and the distance between them is millions of kilometers. Phew - brain hurting yet?
The legal definition of space is 50 kilometers up, but for our purposes, "space" starts about 200 kilometers up.
There's still a heck of a lot of atmosphere ("lot" being a relative term) there, but an average satellite can stay in orbit for weeks before the atmospheric friction slows it enough so it falls out of orbit.
I don't want to belittle what is in todays terms simply a herculean effort. 748 days is 748 days, simple as that, and it's the current record, and more power to Sergei Krikalev frankly.
But at that rate, and we're talking a mere couple of years really (2.04 years), that's not even a one way trip to some of our nearest neighbors, let alone the return trip, if you were thinking of coming back that is * grin *
We are though using developments from space technology to do some pretty weird things, for example, the industrialization of a NASA-tested concept for artificially creating meat!
According to a recently published academic paper, Edible Meat Can be Grown in a Lab on Industrial Scale. ( read more about it at GizMag ).
I wonder if vegies and vegans will now consider eating meat if it doesn't involve killing a living breathing animal? Doubt it.
Anyway - Three cheers for Sergei, but as for the rest of the space program, no dice, we're clearly generations away from truely being a space faring people.
Mores the pity!
It's been a long time coming, but finally it looks like Telstra will be split in two very seperate companies.
One to provide Wholesale Telecoms and the other to provide Retail Telecoms services.
This is so horribly over due!
We may be on the verge, finally, of seeing some real competition in the Australian teleco market place.
Telstra naturally are not very happy about what is now an approved [by cabinet] package, which will force them [Telstra] to create seperage network and retail divisions, with separate premises and management but under the same company structure.
If all goes to plan, the government (coalition) could steamroll ahead and sell its majority share holding ( 51.8 per cent stake ) in Telstra before the end of next year. Something which has been high on the governments agenda for some time now.
Sure that's going to make them a truck load of money, it's (the sale of 51.8% of Telstra is a lot of share value) going to do that no matter how you look at it.
What does all this mean for Australia?
Well we could perhaps look just accross the Tasman to the windy city of Wellington, in New Zealand, for an example of just what a completely deregulated teleco marketplace can do if you allow it.
Businesses and individual end users in Wellington, can gain access to Data and Voice services that the rest of the world ( except perhaps for Singapore with their Interent Corridor ) dreams of.
With 10 megabit and 100 megabit, and even gigabit connections for tens or hundreds of dollars a month, zero data usage charges, and peering for one and all if you want it, Wellington has shown that a completely deregulated telecommunications industry can work, and will work, if you allow it to.
We won't in the near future, say the next five years even, see the likes of what NZ has been able to achieve here in Australia, well not from what I can see gazing into my crystal ball anyway, as there's a legacy culture to be left behind before we can see Australia make major leaps forward.
I'm hoping that with Telstra now having to form a legitimate Wholesale arm, freed up and allowed to sell outside of it's previous one and only customer, being Telstra itself ( oh and the occasional carrier and ISP when they had time of course ), Telstra Wholesale may be allowed to sell core services at prices that would allow 3rd party providers, in particular the DSL providers, or the Broadband market at least, provide now ADSL 2+ services of 22 megabit speeds, throughout the country at prices equal to what we now pay for 1.5 megabit links.
What does the general media have to report? Here's a few links for further homework on the topic:
Let's hope that with new management, and a sense of responsibility to the nation, the new Telstra's can both give back a little of what they have so easily come by, and finally deliver on the government's Digital Nation promise.
More to come on the temp home of Dez's Blog at http://mosman.no-ip.com
Yea, sorry - didn't think my first reply worked and tried to put it in the main area, need more practice!
Dez
Ok,
;-)
I'm going to clock in here with my experience to date with m0n0wall which has been fantastic ( no I don't own shares in anything to do with m0n0wall *grin* - wish I did !! ).
I have to say that from my experience to date with it, m0n0wall is without a doubt one of, if not THE, leading firewall platforms currently available in the open source world, and it's fair to say that I've had a thing or two to do with firewalls and security in general over the past 20+ odd years.
with years and years of hands on design and implementation using checkpoint on sun, checkpoint on nokia, cisco routers, cisco pix, netscreen, ipf, ipfw, iptables, blah blah.
heck, I had such a hard on for checkpoint that at one stage I've even run up a SOFAware box which has the checkpoint inspection module in it, although it's web interface is crap and you can't actually do anything with the firewall policy other than port mapping and translations.
anyway the bastard thing kept resetting and or just slowing down to the point of being so useless I threw it away - after putting it through a hammer test - hammer won * grin *
so I've played with firewalls ok, and god knows how many other bloody firewall platforms, I've played with as many open source firewalls as I can get my hands on, and m0n0wall in particular really has impressed me. When I say play by the way, I mean I've put it through some horrible lab testing, really pushed till smoke came out of the things!
note: firewall blog with reviews of the various firewalls pending kids
smoothwall in my experience had made some very serious inroads towards what was going to become a very strong contender, but then the group fell into ( from what I could tell from the sidelines ) a political infighting jihad which still effects the project.
add to this that they [in my opinion] seemed to have also very seriously stuffed up with their DSL support in 2.x by only supporting USB models of the more widely used DSL modems, particularly here in Australia where Alcatel Speedtouch modems are used far and wide.
in fact it was during an upgrade attempt from smoothwall 1.x to 2.x, I found this out when I was trying to get my DSL modem to talk to smoothwall etc, and out of sheer frustration I decided it was time to dump smoothwall and have another look around.
for a time I even tried running iptables on linux, using fwbuilder on my mac natively and seriously hardened redhat 7.3 ( lord knows it needed it ), horribly stripped down with just enough of the base os left to support two ethernet cards, iptables, and ssh ( to allow fwbuilder to install it's policy ), and I'm still a very big fan of this model, but the one thing that I found a headache setting up and maintaining using fwbulder in this sort of architecture was vpn connections / clients. Also shaping traffic wasn't really feasible and nobody in their right might these days ( again my personal opinion ) runs anything on a network without some form of shaping! Do they?
so again I went hunting the open source tundra for a new toolset. this was when I re-discovered m0n0wall, which when I first reviewed it, was perhaps at a very early stage in it's life cycle and by no means the magical wonderland that it is todya [as of 1/6/2005 (that's July 1st for you American date centric folk)].
Key strengths that I've had working and under high loads, include:
- base firewall policy made up of some very complex rules
- multiple dmz's ( I hate dmz's - they are lame but so be it )
- nat on wan interface, and one of the dmz interfaces
- multiple static routes
- multiple dynamic routes
- dynamic dns ( had to tinker to get no-ip.com working but hey )
- dns caching / forwarding
- ipsec and pptp vpn connections with many vpn clinets
- traffic shaping with QoS which actually works! yea, it really does!
- address aliases on floating ip's for fail over / redundancy
- dhcp with pool of ip's as well as fixed MAC map's and static ip's
- proxy
Ok,
;-)
I'm going to clock in here with my experience to date with m0n0wall which has been fantastic ( no I don't own shares in anything to do with m0n0wall *grin* - wish I did !! ).
I have to say that from my experience to date with it, m0n0wall is without a doubt one of, if not THE, leading firewall platforms currently available in the open source world, and it's fair to say that I've had a thing or two to do with firewalls and security in general over the past 20+ odd years.
with years and years of hands on design and implementation using checkpoint on sun, checkpoint on nokia, cisco routers, cisco pix, netscreen, ipf, ipfw, iptables, blah blah.
heck, I had such a hard on for checkpoint that at one stage I've even run up a SOFAware box which has the checkpoint inspection module in it, although it's web interface is crap and you can't actually do anything with the firewall policy other than port mapping and translations.
anyway the bastard thing kept resetting and or just slowing down to the point of being so useless I threw it away - after putting it through a hammer test - hammer won * grin *
so I've played with firewalls ok, and god knows how many other bloody firewall platforms, I've played with as many open source firewalls as I can get my hands on, and m0n0wall in particular really has impressed me. When I say play by the way, I mean I've put it through some horrible lab testing, really pushed till smoke came out of the things!
note: firewall blog with reviews of the various firewalls pending kids
smoothwall in my experience had made some very serious inroads towards what was going to become a very strong contender, but then the group fell into ( from what I could tell from the sidelines ) a political infighting jihad which still effects the project.
add to this that they [in my opinion] seemed to have also very seriously stuffed up with their DSL support in 2.x by only supporting USB models of the more widely used DSL modems, particularly here in Australia where Alcatel Speedtouch modems are used far and wide.
in fact it was during an upgrade attempt from smoothwall 1.x to 2.x, I found this out when I was trying to get my DSL modem to talk to smoothwall etc, and out of sheer frustration I decided it was time to dump smoothwall and have another look around.
for a time I even tried running iptables on linux, using fwbuilder on my mac natively and seriously hardened redhat 7.3 ( lord knows it needed it ), horribly stripped down with just enough of the base os left to support two ethernet cards, iptables, and ssh ( to allow fwbuilder to install it's policy ), and I'm still a very big fan of this model, but the one thing that I found a headache setting up and maintaining using fwbulder in this sort of architecture was vpn connections / clients. Also shaping traffic wasn't really feasible and nobody in their right might these days ( again my personal opinion ) runs anything on a network without some form of shaping! Do they?
so again I went hunting the open source tundra for a new toolset. this was when I re-discovered m0n0wall, which when I first reviewed it, was perhaps at a very early stage in it's life cycle and by no means the magical wonderland that it is todya [as of 1/6/2005 (that's July 1st for you American date centric folk)].
Key strengths that I've had working and under high loads, include:
- base firewall policy made up of some very complex rules
- multiple dmz's ( I hate dmz's - they are lame but so be it )
- nat on wan interface, and one of the dmz interfaces
- multiple static routes
- multiple dynamic routes
- dynamic dns ( had to tinker to get no-ip.com working but hey )
- dns caching / forwarding
- ipsec and pptp vpn connections with many vpn clinets
- traffic shaping with QoS which actually works! yea, it really does!
- address aliases on floating ip's for fail over / redundancy
- dhcp with pool of ip's as well as fixed MAC map's and static ip's
- proxy
I think those of us with Mac OS X and half a klew already use standards compliant methods to make sure scripts work on any platform, i.e.:
a rch.COM.AU
#!/bin/sh -
echo "Hello World!"
exit 0
99.2% of people using Mac OS X prolly don't use a shell anyway.
With 18+ years unix I've gone from hard copy tty line printer consoles, to vt100's, wyse 75, wysel 180, to a cranky old toshiba 5200 plazma screen ( loved that ) running Xenix, then SCO, then 386/ix, and eventually BSD/386. I tried using a toshiba techra 8200 with windows 95, 98, 2000 and it was crap, then went to using an old Mac Powerbook 150 and never looked back - I've always had a mac around the place, but once I started using a mac, I found it the closest thing to having a sun workstation in a laptop, without the cli ( bugger ) - I even ran Apple's "Apple UX" BSD system once, but it was heavy on old mac's, even ran the tenon stuff - same deal.
So when Mac OS X 10.0 came out, I dropped everything and ran to the nearest ( yes, I said ran! ) shop, and got a 15 inch G4 and have never looked back. Now I can't understand why people run PC platform latop's with Linux when they can have the perfect world of a Mach kernel ( the best ), a FreeBSD core with the whole of the open source world at their feet, and everything I have ever needed that ran on Linux runs on Darwin just fine ( unless some moron has written pissy code and then they can get stuffed - I'll write it myself ).
Convert now - dump intel, bring your linux thinking, and run Darwin on PPC and see a brighter world!
++dez;
---
Dez Blanchfield
dez@blanchfield.com.au
http://WebSe