Note the word CONGRESS, it does not say anyone else can not, it just says that congress can not. I can not cuss out my boss and be protected, I CAN cuss out my government though.
Re:I may be an old fart but...
on
IETF vs. ICANN
·
· Score: 2
What good it would be is to add a hopefully "descriptive" tag to the URL. If it's got a.net then it's most likeley a ISP, since a commercial company is most likely going to use.com
I'm not going to go much back over in what I've replied to the statement about flattening of the namespace... but essentially nobody is doing what you're saying, when a company registers a domain they are registering every possible combination of tld's they can get... effectively flattening the namespace, making the tld non-descriptive, and just giving money to the registers for no good reason.
Once people actually *have* to make a choice and decide where they should be, would adding additional truely descriptive tld's be effective. Since there would actualy be purpose to the tld.
Re:I may be an old fart but...
on
IETF vs. ICANN
·
· Score: 3
Look at the way it is today, if everybody was forced to use only *.com whoever had a.net,.org, etc. just lost their domain; would the internet only have a third of it's servers??? No, most likely 90% of the servers would still be there, because companies are not just registering one tld they are registering all they can get their hands on.
Using your example, look at slashdot, there's a.com &.org, going to the same place, slashdot is stepping on two-thirds of the space already. Right now tld's are completely useless: they add confusion to the consumer (or websurfer), make me type extra characters for no good reason, currently they're only really good for lining registers pockets with money by having the same company register the one domain 50 billion different ways.
If we did this there would actually be a point to adding additional tld's since you'd sit in whichever one made the most sense; instead of getting every one you can get your hands on, no matter if it made any sense at all.
And just be a bastard, why don't you just include that tld information into your domain??? slashdotcom.com there you have your two-thirds of domains back:)
Re:I may be an old fart but...
on
IETF vs. ICANN
·
· Score: 2
Ok, but ask yourself this; is motor parts joeblow registering just joeblow.com, and NOT joeblow.biz/org/net/etc? You can be if he has the chance he's going to register all tld's that he can.
When a company currently registers a domain it jumps on as many tld's it can get, making the point of tld's completely useless. I'm just suggesting that typing the extra tld in my url to actually be useful again, instead of just being useful to the registers to make money off of.
I may be an old fart but...
on
IETF vs. ICANN
·
· Score: 5
My opinion on the whole TLD mess is that there should only be one domain.[net|org|com|.*] Instead of having slashdot.org & slashdot.com that whoever registers it has to pick a tld and the rest of the tld's with that domain are unavailable.
What this would get me is the confidence that I can say a.net is most likely a network provider, a.org is an organization, etc. Since you'd pick which tld is going to be most appropriate for you business.
Also we don't have lawsuits between joeblow.biz & joeblow.com, etc. since they couldn't exist because they'd have a domain clash.
While I'm on a roll (rant), I'd even like it better if you actually had to show you truley belonged in a certain tld (you have to provide network services to be in the.net tld).
I have no idea how to reverse the mess with all the different organization that are in the mess of having conflicts, but I know it could be implented for any new domains.
Cool, I was a bit harch earlier and should have probably, should have replied immediately have *that* meeting I had.
One of the better places to go if you don't have the freeware CD's is the freeware.sgi.com website. It includes gnome (albeit it can be a big old pain finding all of the damn dependencies one by one, dependencies requiring dependencies). Lots of good stuff out there (it's same that's on the freeware CD) it's for anything 6.2 onward.
As a FYI, on the installation part, the best way I've found on dealing with it is to copy all of your 6.5 base CD's up to a nfs partition, and you can multiple CD's into the same directory. You don't have to worry about switching CD's because you can't find a component, we then have a directory for each of the overlays (6.5.1, 6.5.2). I think I know where you're coming from on your conflict pain, this is where having the nfs distribution works well, open inst with your / then do an open with / , do a "keep *" then install which ever piece of software you need, works good for me.
I probably should have clarified what I was saying about the security thing a bit, I was making more of a point as to how things were and how things are now. On 6.5 they've introduced "privileges" which is an admin gui that allows one to give rights to specific users to do things (add printers, mount filesystems, etc.), almost a "sudo" type of a thing, works well for users who might need to bring up a nfs mount now and then, but that it. But what 6.5 really gets you is ACL's which came from their Trusted Irix software, which if you want to get nitty gritty will do some nice stuff. I think ipfilterd existed with 6.2 (I could be wrong) which is effectively a very fast in kernel network acl filter, it's not stateful, but has some very neat features.
I understand about the availability of binaries, we've ran into it more than a couple of times on Irix (actually had people use our system to build binaries for them). SGI is using Linux as their way out on this one, people won't develop for Irix, well people are developing for Linux...
To be completely honest your remark on stability... well all I can say is stability normally comes from the admin. I'm not going to get into a pissing match about uptime, but neadless to say mine have been up in heavy use for a long time. I've got hundreds of SGI systems up with no problems at all.
Having cut my teeth on a good old CRDS (Charles River Data Systems, PDP11 clone). I found moving to Irix no problem at all, easier than going from SunOS to Solaris. Yes you had to have some intelligence in admining your box, you couldn't just "point and drool"; but then again I don't know any admin that uses the gui admin apps on Irix, Sun, etc. We've stopped installing the vmsa (formerly vxvm) veritas volume manager because we do everything through a command line.
Security... well you do have a point there, but again it's what you install. Look at a default Sun install, there's another security nightmare waiting to happen. SGI's 6.5 release years ago really cleaned up this mess, for the past 3 years they've beaten Solaris on number of security vulnerabilities (Bugtraq Vulnerability stats from securityfocus.org), and have yet to have one this year. With 6.5 they added all the nice security features, ACL's, priviliges, etc. they got their sh*t together unlike almost every other OS out there. They even have a "Improve System Security" option which does all of the normal admin hardening work for you (even get's down to who can run javascript in Netscape on the system)
I don't know where you are coming from on the installation thing, for years Irix has had NFS, before that they had to pay royalties to Sun (which is why you had to pay for it). The dependencies thing can be kinda weird, but I'd say it's easier than anything else out there, it tells you exactly what package you need, unlike running "rpm -i" and it just spits out the file you need, no package information.
Compatibility??? I've got an old Indigo II that I've got almost everything 3rd party: ram, all hard drives, cdrom, dat drive. The only that was gotten specific for this system was the ram, hard drive came out of an Sun system, CDROM was from a PC and the dat from a Sequent (it's a jukebox changer even). It all worked as soon as I plugged it in (except for the changer, it was unable to read some specific information about the drive so I had to tell it what it was). The only time I've had a problem was when someone handed me a differential drive in a single ended case and I plugged it in, it obviously couldn't see the drive. CDE why don't you just ask for it??? It's available, but nowadays almost everybody here uses GNOME or 4DWM which are on the CD that come with the system (along with NFS that you can't seem to find).
Maintainability, 6.3 was meant to specifically support the O2, 6.4 came out with their Origin line to support the Origin hardware. 6.5 brought the hardware support back inline amongst all system. I'm not saying it wasn't painful but, again that was multiple years ago to support their new hardware. The splits were just to support certain NEW hardware that hadn't been around before.
Support, I've heard both good and bad things from different people. We have excellent support, better than any other vendor (including EMC and that's their entire claim to fame). I will say this though, EVERY company I've talked to that has had both SGI & Sun have said that Sun support has seriously sucked in comparison. I'm in a much larger organization, that purchases millions of dollars of SGI equipment you tend to get a different bread of support (along with any organization)... But when I was a lone admin with 2 SGI boxes, sitting a 5 hour drive from the nearest SGI tech, I still got good service. I didn't have a SGI's ear for enhancements, etc. but I had good support.
From what I can tell most of your pain has come from years ago, 6.5 was released 3 years ago. If you want to start comparing things that long ago, I could drag up some painful moments from using SunOS/Solaris or Ygdrassil & Slackware Linux and do some comparisons too.
This could be extremely BAD, think about if someone were to add, modify just a couple of lines of code. As long as people have been keeping original, master copies offsite it shouldn't be too dificult to get things back to normal.
But could you imaging though, if someone were to add/modify 4-5 lines off code and you didn't have another copy offsite? You would have to remember why every single line of code was there, you couldn't trust the comments telling you why this procedure is there.
For an example: It would be very easy to add code to add, "allow-anyone with this password to get a root shell" lines to a compromised box. Now lets say it's an ISP running an IMAP server they get off of Sourceforge, gets a new version. The program appears to run just like it's supposed to, they never would know that there is a 3 line, fork shell process code added into it. Being an ISP they have to have this program accessible to anyone anywhere, so they can't have this program behind a firewall. Along comes the wiley hacker who compromised the code to begin with... *poof* root shell on box.
That is the scary part, without being able to go back to a true "golden" state, EVERY single line of code has to be checked, without relying on possibly forged comments. Comparing MD5 hashes could tell you if the program has been modified, but some programs are modified hourly, most of these don't have checksum information.
I beg to differ, write cacheing is normally ENABLED by default. If the drive was in full sync, then your performance would be complete crap. Everytime you did anything you'd have to wait for the data to get destaged onto the drive (very painful).
Do a "dd if=/dev/zero of=./file.test", wait a bit and break out of it (you could also do a rm -rf on a larger directory with lots of files). It will spit out how much it has supposedly written out to the drive, then quickly do a sync, you'll have to wait a bit while the data gets out of cache. The more memory you have the more pronounced this becomes, the sync destages the data to the drive from write cache. My box with 2gig of ram "supposedly" wrote ~500mb out to a single drive in just about 5 seconds... until I typed sync.
They way I see it is that Reiser is looking to make a mark with having the latest & greatest feature list, XFS is looking to have the the most stable & reliable filesystem, and ext3 is a way to have people keep their existing ext2 data and add journaling support.
Reiser is planning on selling their modules in the future, make a new feature to be sold and change the previously sold module to be free. Their entire business model depends on them having newer and newer features, which is great for people who are wanting/needing feature over stability.
XFS is leaning more towards the datacenter type of situation, it may not have the latest and greatest, but it will work reliably, constantly, and with great performance. XFS is looking towards Linux as their OS platform, they have to give the same quality of filesystem they had on Irix to their customers who demand that quality. (when buying a multi-million dollar 512proc numa system they tend to require lots of stability).
On the competeing filesystems, Steve Lord from the XFS mailing list said it probably best:
"...I have never regarded the different filesystem on Linux as being in direct competition with each other, there will always be benefits to using each different filesystem for their strong points. Plus having several filesystems under active development means that there will be a tendency for the developers to make theirs the best, the implementations improve, and everyone wins."
Reiser will be used for the things it's good at (squid, mail spool, new features) and XFS will be used for the things it's good at (larger files, NFS server, stability). They compete only that they are filesystems, but what they are designed to be good at are two different things.
Only problem is that all of the SSD's I know of are drive backed up with a spinning disk. In the event of power, etc. loss they start destageing all their information to spinning disk. When they get turned backon it takes 20 minutes or so to get fully back off of the drive.
Re:2600 is completely WRONG on this one.
on
2600 v. Ford Motors
·
· Score: 1
The counter argument is: does the driver of said car need to stop the car and look at the manufacturer information on every street sign to make sure that the direction information comes from the correct place?
My belief why NASA tends to very opposed to this type of thing is that they are scared sh*tless that they are going to lose funding because of it. All it's going to take is someone to get the bright idea that instead of space researched paid by taxes, that it could completely survive on the private sector. Think if Congress started siphoning off the budget money, and say "if you want some more money throw up some tourists, to fund your next deep space satelite". Funding from the next solar panel, will come from the 5 tourists that have to be shot up, because Congress decided they didn't want to give us any more money.
I don't think NASA is all that concerned about someone going through the proper training, etc. and being a safety hazard, but more about what this could truely mean, to all those nickles and dimes they've had to beg and plead for. Could cause lots of problems on actually getting true research done.
If you compare prices on a O3000 compared to a Starfire there prices are pretty clueful
Service/support, we have great support from SGI (of course we have lots and lots of servers). From their database side, what I had heard is that someone at SGI pissed off Larry Ellison at Oracle, so he's let SGI whither on the vine.
Poor commitment, yes when the bastard (who I shall not say) went to Microsoft was running things. I almost think he put the 320 & 520 into SGI's line just to get a job at Microsoft.
Which technology... for the 3000 we've got here I can get SCSI, FC, PCIx, Myranet, Gig E, OC3, FDDI, USB, Firewire, plus all of the special digital video stuff you want.
Longevity... no question here
Marketing has been a thorn in SGI's side ever since I've used them. Their Origin line was a kick ass server at the time, but everybody only thought of them as makers of movie effects (and still do to an extent)
The problem becomes from a price/performance metric as to what database do you use? The DB2 software itself was more expensive then the servers or the storage. 1/3 of the price of the entire system (300k+) just for the database software tends to eat into the price quickly, but I don't know of an alternative. When we checked out Oracle for Linux it was almost $200k for a single Compaq 2x PIII box, about the only option they had were MySQL or Postgress to reduce costs which would not have been up to the task.
Normally you don't want to make your filesystems that big anyway, from a backup/restore issue it becomes a killer. Any DBA worth his salt uses raw filesystems if possible, meaning you can only have one backup process on that partition i.e. normally meaning only one tape drive spinning for the entire partition. Retrieving from a single tape we can get 10-15 gig/hour, now for a restore of a 2 terabyte partition you'd be waiting a LONG while, before you could even think of applying your redo logs.
They have given an amazing level of commitment to IA32, but they have given even more to IA64 which will be very interesting IF they can pull it off. IA64 Linux numa clusters scalling to 1024 proc as a single image... that's what I call sweet.
I can deffinetly say that I like what SGI has done with the O3k series, it's cheaper than the O2K, it's a bit faster with the same mem & cpu (the faster interconnects give the boost), and I can just add an additional "p-brick" if I need more processing power without having to purchase all of the chassis stuff (XIO, disk spaces, etc). Not to mention the system partitioning capability, which I can't believe SGI slip through there fingers by allowing Sun to purchase the tech from Cray when they bought them... but that was when the Microsoft bastard was at the helm.
And a loom only *this* big, operated by swarms of bluecollar crickets. Working perfectly until they go on strike, wanting 4 weeks of paid vacation and free nectar services; throwing the production schedule of the elevator months behind schedule and overbudget.
I would say that trademarks are meant exactly for the purpose you state in the last paragraph.
Just because it is opensource doesn't make it any different than the rest of the world. I can't make a company "Tosheba" making a product named "50HD70" looking almost exactly like Toshiba's 50HX70, and expect not to have problems. If they were products who DID NOT compete no problem; but the way you put it, you are riding the coat-tails of the commercial product introducing ambiguity. Why should you want to name a competeing product the same as the commercial, except for to get the name recognition from the commercial product, which again is what trademarks are for.
If the opensource product has merits it should be able to stand on it's own, and as you say "compete" with the commercial product; hence it should not even have to have a name close to the commercial products. Look Gimp & Photoshop, the names aren't anywhere near the same, but they tend to compete against each other, and Gimp did it on it's own without having to bootstrap itself up from the name recognition of someone elses trademark.
Spelling & grammar checker off because I don't care
Max Filesystem size: 18 million TB
Max File size: 9 million TB
That's according to their spec sheet, I could only dream I had 18 exabytes, course then I might need something bigger than an Indigo II, to get good use out of it.
I asked this of our rep a few months ago, asking about the port work of Linux to Mips. What he said was that Irix has a buch of special magic going on in conjunction with the Mips CPU. i.e. make sure code gets to the cpu through at the same time, special read ahead things, tweaks with the bus and cpu, etc. to put them into a Linux system would fundamentaly change how things are currently done in most of the kernel . Think how hard they've been trying just to get big memory support into the kernel, could you imaging trying to get a completely different core change that would only help Mips cpu's.
They are doing porting work, and Linux is running on Mips currently (so is NetBSD), but the special performance enhancements aren't in the kernel. Currently running Irix on a Mips Origin will be much faster than with Linux, but on the smaller workstations O2, Indigo II, etc. Linux will be faster (for non gui work) because it's so light-weight.
Don't plan on getting a speedy Linux on your Onyx, Origin systems anytime soon, unless you want to take a big performance cut.
The key word with the law is *ALLOWS*. Not being familiar with the law, but can make a reasonable stab at it from your statements. It doesn't make it illegal to make a copy of the CD's, but it at the same time doesn't say that the producer has to make it easy for you to make a copy.
I guess their reaction would be no problem. If you've been keeping up with SGI, they have been doing major things with Linux. They are selling Linux systems with excellent support (not to mention 24x7 hardware & OS support was less then IBM or Compaq by thousands of dollars). With SGI licensing their tech to nvidia if they don't use a SGI system, they are probably using a nvidia card and get a cut their. The nice thing is when someone goes, your Irix workstations are too expensive but we want your O2k server, SGI just says why don't you look at our Linux workstation with a tweaked up nvidia card that smokes everyone else.
Later this year when IA64 is hopefully released they'll roll out their Intel 3400 series line, which can scale to 512 proc's running Linux without clustering (love that numa). They currently have all of the IA32 bit offerings requested 1-4CPU small form factor servers. Once they get the IA64 goin SGI will be able to offer Linux workstations to 1U clustered servers (which they are selling today for beowulf), all they way up to a monster 512 proc numa cluster (talking with our SGI rep yesterday, someone over in the Netherlands purchased 8x 128 proc mips systems, that they then cluster to make a 1024 proc system with failover between nodes).
I feelings on the apache project things, is that they've tried unsuccessfully for so long to get the changes into the standard build, that someone said we've got to: get some traction and get it into the build, fork it into our own distro, or concede and reallocate the expensive people we've got working on it. I think SGI started to get the feeling that they were beating a dead horse.
Having just gone through a fairly rigorous review of the possiblity of using Linux for our production environment, a couple of things I've found.
NFS works, but can cause lots of grief (file handle cache, etc.)
Their are apps that cause problems with ReiserFS (Qmail has to be patched)
Root filesystem is pretty iffy
None of the journaled filesystems support nfs locking, also using a journaled fs on the root partition is pretty much use at your own risk for all of the fournaled fs at this time. Qmail has some issues with synchronous tasks with ReiserFS (FAQ at Namesys explains this)
People are throwing up ReiserFS in front of me, when I say production Linux with lots of storage of storage (we deal in lots of terabytes here...) is a "no" right now. I just throw back that if I have to worry about the app not working with my filesystem and by the way don't do any writes over NFS, then I don't feel comfortable running it in production (not to mention that nasty root compromise with ReiserFS just a few weeks ago).
Don't get me wrong I'm running both XFS & ReiserFS currently and like both, but they ain't getting anywhere near my production EMC storage, until they grow up a bit more. (on a side note we are using Linux in production, but only on things with 50 gig of storage or less, and can sustain some downtime, with limited revenue loss)
Your model assumes that the backup would be able to restore to the exact moment people were working at. We have over a hundred developers here, and if something bad were to happen to our devel environment (admins had a pc-nfs mount somewhere bad), if we where to throw out half the developers (for as you say breaks, etc) and luckily be out only 4 hours from the last backup... (which actually tends to be 8 hours for most places, others do it hourly, or every other hour)
50 workers * 4 hours = 200 lost man hours just for the developers to replace what they did since last backup + however long developers have to sit on their thumbs while the restore occurs + all the rest of the things (time diverted from other projects, etc.) you end up with a lot of time.
Your model mimics the old joke email 260 million people in the US, 114mil retired, 93 mil in school... 206,000 in hospitals... leaving only one person doing all the work.
Re:Good, The New Workers need to unionise.
on
The Jungle
·
· Score: 5
Being in the shadow of the biggest concentration of unions (Detroit's "big three" auto industry), I would say you are completely wrong. A union is kinda like Communism, it looks great on paper but it never seems to work out like it did on the big chief notebook.
Case in point, in Ford one cannot move a monitor on ones own desk (move it over a foot to adjust for glare, etc) you have to have one of the "union" guys do it. If you do move your monitor, you get written up and the company pays the union worker for the "work" that you did... just to justify that some dumbass actually requires a job. The tech guys are not allowed to put the $50k cad workstation they built on an engineers desk they have to get a "union electrician" to plug the cables in. How completely idiotic is this? They build it, installed it, but they can't plug it into a 110 volt outlet in the wall. What they would actually do is, tell the union guy how much the workstation cost (do you want to be responsible for it, scares most union guys), and if he would let them do the work, the union guy could take credit and the pay for it. Useless absolutely useless, all from the outgrowth of unions, and the belief that because you hired a dumbass, you should never get rid of dumbass screwoffs. (Don't even get me started on the union seniority bullshit, how old you are, not your performance determines how much you get paid... ugh).
Spelling and grammar checker off because I don't care
The problem with that (and it's a very well known problem) is that it doesn't represent the country, it represents specific portions of the country; and if you think about it, it's symbolic of special interest groups.
The electorial college gives the states more power, so as to prevent candidates from just campaining in the 10 or so most populous states and essentially concede their every whim to them and screw the rest of the states of the country. Having the electorial college, actually gives the candidates reason to try and do what's best for the entire country, not just certain portions of it. This way the small guy's voice actually is paid attention to, how would you like it if you lived in somewhere less populous and because the vote was based entirely on the popular decision the candidates went to Cal & NY and said everyone's taxes in Montana, Wyoming, Kansas, etc. will go to paying Cal & NY taxes so they won't have to pay any this year because they are going to vote for me... now think about not being able to do a damn thing about it because the 10% of states they're giving tax breaks to have more votes than the rest of the 90% combined.
The candidates are responsible for the nation as a whole not to just kiss the ass of who's got the most people living in certain portions.
Spelling and grammar checker off because I don't care
The previous poster didn't state it very well, but his general "gist"(sp?) was correct.
To use your analogy, being attacked with a hammer is not categorized as assault, but assault with a deadly weapon; the hammer is not illegal to use, but to use it for assault, etc. is. That's what this treaty is trying to do, you use a tool in a manner that is against the law, your mis-use of the tool can be used against you.
Slashdot Mirror
Umm... YOU are missing the point.
From your quote
"Congress shall make no law...."
Note the word CONGRESS, it does not say anyone else can not, it just says that congress can not. I can not cuss out my boss and be protected, I CAN cuss out my government though.
What good it would be is to add a hopefully "descriptive" tag to the URL. If it's got a .net then it's most likeley a ISP, since a commercial company is most likely going to use .com
I'm not going to go much back over in what I've replied to the statement about flattening of the namespace... but essentially nobody is doing what you're saying, when a company registers a domain they are registering every possible combination of tld's they can get... effectively flattening the namespace, making the tld non-descriptive, and just giving money to the registers for no good reason.
Once people actually *have* to make a choice and decide where they should be, would adding additional truely descriptive tld's be effective. Since there would actualy be purpose to the tld.
Look at the way it is today, if everybody was forced to use only *.com whoever had a .net, .org, etc. just lost their domain; would the internet only have a third of it's servers??? No, most likely 90% of the servers would still be there, because companies are not just registering one tld they are registering all they can get their hands on.
.com & .org, going to the same place, slashdot is stepping on two-thirds of the space already. Right now tld's are completely useless: they add confusion to the consumer (or websurfer), make me type extra characters for no good reason, currently they're only really good for lining registers pockets with money by having the same company register the one domain 50 billion different ways.
:)
Using your example, look at slashdot, there's a
If we did this there would actually be a point to adding additional tld's since you'd sit in whichever one made the most sense; instead of getting every one you can get your hands on, no matter if it made any sense at all.
And just be a bastard, why don't you just include that tld information into your domain??? slashdotcom.com there you have your two-thirds of domains back
Ok, but ask yourself this; is motor parts joeblow registering just joeblow.com, and NOT joeblow.biz/org/net/etc? You can be if he has the chance he's going to register all tld's that he can.
When a company currently registers a domain it jumps on as many tld's it can get, making the point of tld's completely useless. I'm just suggesting that typing the extra tld in my url to actually be useful again, instead of just being useful to the registers to make money off of.
My opinion on the whole TLD mess is that there should only be one domain.[net|org|com|.*] Instead of having slashdot.org & slashdot.com that whoever registers it has to pick a tld and the rest of the tld's with that domain are unavailable.
.net is most likely a network provider, a .org is an organization, etc. Since you'd pick which tld is going to be most appropriate for you business.
.net tld).
What this would get me is the confidence that I can say a
Also we don't have lawsuits between joeblow.biz & joeblow.com, etc. since they couldn't exist because they'd have a domain clash.
While I'm on a roll (rant), I'd even like it better if you actually had to show you truley belonged in a certain tld (you have to provide network services to be in the
I have no idea how to reverse the mess with all the different organization that are in the mess of having conflicts, but I know it could be implented for any new domains.
Of course that's me being an old curmudgeon
Cool, I was a bit harch earlier and should have probably, should have replied immediately have *that* meeting I had.
One of the better places to go if you don't have the freeware CD's is the freeware.sgi.com website. It includes gnome (albeit it can be a big old pain finding all of the damn dependencies one by one, dependencies requiring dependencies). Lots of good stuff out there (it's same that's on the freeware CD) it's for anything 6.2 onward.
As a FYI, on the installation part, the best way I've found on dealing with it is to copy all of your 6.5 base CD's up to a nfs partition, and you can multiple CD's into the same directory. You don't have to worry about switching CD's because you can't find a component, we then have a directory for each of the overlays (6.5.1, 6.5.2). I think I know where you're coming from on your conflict pain, this is where having the nfs distribution works well, open inst with your / then do an open with / , do a "keep *" then install which ever piece of software you need, works good for me.
I probably should have clarified what I was saying about the security thing a bit, I was making more of a point as to how things were and how things are now. On 6.5 they've introduced "privileges" which is an admin gui that allows one to give rights to specific users to do things (add printers, mount filesystems, etc.), almost a "sudo" type of a thing, works well for users who might need to bring up a nfs mount now and then, but that it. But what 6.5 really gets you is ACL's which came from their Trusted Irix software, which if you want to get nitty gritty will do some nice stuff. I think ipfilterd existed with 6.2 (I could be wrong) which is effectively a very fast in kernel network acl filter, it's not stateful, but has some very neat features.
I understand about the availability of binaries, we've ran into it more than a couple of times on Irix (actually had people use our system to build binaries for them). SGI is using Linux as their way out on this one, people won't develop for Irix, well people are developing for Linux...
I've had a complete opposite reaction with Irix.
To be completely honest your remark on stability... well all I can say is stability normally comes from the admin. I'm not going to get into a pissing match about uptime, but neadless to say mine have been up in heavy use for a long time. I've got hundreds of SGI systems up with no problems at all.
Having cut my teeth on a good old CRDS (Charles River Data Systems, PDP11 clone). I found moving to Irix no problem at all, easier than going from SunOS to Solaris. Yes you had to have some intelligence in admining your box, you couldn't just "point and drool"; but then again I don't know any admin that uses the gui admin apps on Irix, Sun, etc. We've stopped installing the vmsa (formerly vxvm) veritas volume manager because we do everything through a command line.
Security... well you do have a point there, but again it's what you install. Look at a default Sun install, there's another security nightmare waiting to happen. SGI's 6.5 release years ago really cleaned up this mess, for the past 3 years they've beaten Solaris on number of security vulnerabilities (Bugtraq Vulnerability stats from securityfocus.org), and have yet to have one this year. With 6.5 they added all the nice security features, ACL's, priviliges, etc. they got their sh*t together unlike almost every other OS out there. They even have a "Improve System Security" option which does all of the normal admin hardening work for you (even get's down to who can run javascript in Netscape on the system)
I don't know where you are coming from on the installation thing, for years Irix has had NFS, before that they had to pay royalties to Sun (which is why you had to pay for it). The dependencies thing can be kinda weird, but I'd say it's easier than anything else out there, it tells you exactly what package you need, unlike running "rpm -i" and it just spits out the file you need, no package information.
Compatibility??? I've got an old Indigo II that I've got almost everything 3rd party: ram, all hard drives, cdrom, dat drive. The only that was gotten specific for this system was the ram, hard drive came out of an Sun system, CDROM was from a PC and the dat from a Sequent (it's a jukebox changer even). It all worked as soon as I plugged it in (except for the changer, it was unable to read some specific information about the drive so I had to tell it what it was). The only time I've had a problem was when someone handed me a differential drive in a single ended case and I plugged it in, it obviously couldn't see the drive. CDE why don't you just ask for it??? It's available, but nowadays almost everybody here uses GNOME or 4DWM which are on the CD that come with the system (along with NFS that you can't seem to find).
Maintainability, 6.3 was meant to specifically support the O2, 6.4 came out with their Origin line to support the Origin hardware. 6.5 brought the hardware support back inline amongst all system. I'm not saying it wasn't painful but, again that was multiple years ago to support their new hardware. The splits were just to support certain NEW hardware that hadn't been around before.
Support, I've heard both good and bad things from different people. We have excellent support, better than any other vendor (including EMC and that's their entire claim to fame). I will say this though, EVERY company I've talked to that has had both SGI & Sun have said that Sun support has seriously sucked in comparison. I'm in a much larger organization, that purchases millions of dollars of SGI equipment you tend to get a different bread of support (along with any organization)... But when I was a lone admin with 2 SGI boxes, sitting a 5 hour drive from the nearest SGI tech, I still got good service. I didn't have a SGI's ear for enhancements, etc. but I had good support.
From what I can tell most of your pain has come from years ago, 6.5 was released 3 years ago. If you want to start comparing things that long ago, I could drag up some painful moments from using SunOS/Solaris or Ygdrassil & Slackware Linux and do some comparisons too.
This could be extremely BAD, think about if someone were to add, modify just a couple of lines of code. As long as people have been keeping original, master copies offsite it shouldn't be too dificult to get things back to normal.
But could you imaging though, if someone were to add/modify 4-5 lines off code and you didn't have another copy offsite? You would have to remember why every single line of code was there, you couldn't trust the comments telling you why this procedure is there.
For an example: It would be very easy to add code to add, "allow-anyone with this password to get a root shell" lines to a compromised box. Now lets say it's an ISP running an IMAP server they get off of Sourceforge, gets a new version. The program appears to run just like it's supposed to, they never would know that there is a 3 line, fork shell process code added into it. Being an ISP they have to have this program accessible to anyone anywhere, so they can't have this program behind a firewall. Along comes the wiley hacker who compromised the code to begin with... *poof* root shell on box.
That is the scary part, without being able to go back to a true "golden" state, EVERY single line of code has to be checked, without relying on possibly forged comments. Comparing MD5 hashes could tell you if the program has been modified, but some programs are modified hourly, most of these don't have checksum information.
I beg to differ, write cacheing is normally ENABLED by default. If the drive was in full sync, then your performance would be complete crap. Everytime you did anything you'd have to wait for the data to get destaged onto the drive (very painful).
Do a "dd if=/dev/zero of=./file.test", wait a bit and break out of it (you could also do a rm -rf on a larger directory with lots of files). It will spit out how much it has supposedly written out to the drive, then quickly do a sync, you'll have to wait a bit while the data gets out of cache. The more memory you have the more pronounced this becomes, the sync destages the data to the drive from write cache. My box with 2gig of ram "supposedly" wrote ~500mb out to a single drive in just about 5 seconds... until I typed sync.
They way I see it is that Reiser is looking to make a mark with having the latest & greatest feature list, XFS is looking to have the the most stable & reliable filesystem, and ext3 is a way to have people keep their existing ext2 data and add journaling support.
Reiser is planning on selling their modules in the future, make a new feature to be sold and change the previously sold module to be free. Their entire business model depends on them having newer and newer features, which is great for people who are wanting/needing feature over stability.
XFS is leaning more towards the datacenter type of situation, it may not have the latest and greatest, but it will work reliably, constantly, and with great performance. XFS is looking towards Linux as their OS platform, they have to give the same quality of filesystem they had on Irix to their customers who demand that quality. (when buying a multi-million dollar 512proc numa system they tend to require lots of stability).
On the competeing filesystems, Steve Lord from the XFS mailing list said it probably best:
"...I have never regarded the different filesystem on Linux as being in direct competition with each other, there will always be benefits to using each different filesystem for their strong points. Plus having several filesystems under active development means that there will be a tendency for the developers to make theirs the best, the implementations improve, and everyone wins."
Reiser will be used for the things it's good at (squid, mail spool, new features) and XFS will be used for the things it's good at (larger files, NFS server, stability). They compete only that they are filesystems, but what they are designed to be good at are two different things.
Only problem is that all of the SSD's I know of are drive backed up with a spinning disk. In the event of power, etc. loss they start destageing all their information to spinning disk. When they get turned backon it takes 20 minutes or so to get fully back off of the drive.
The counter argument is: does the driver of said car need to stop the car and look at the manufacturer information on every street sign to make sure that the direction information comes from the correct place?
My belief why NASA tends to very opposed to this type of thing is that they are scared sh*tless that they are going to lose funding because of it. All it's going to take is someone to get the bright idea that instead of space researched paid by taxes, that it could completely survive on the private sector. Think if Congress started siphoning off the budget money, and say "if you want some more money throw up some tourists, to fund your next deep space satelite". Funding from the next solar panel, will come from the 5 tourists that have to be shot up, because Congress decided they didn't want to give us any more money.
I don't think NASA is all that concerned about someone going through the proper training, etc. and being a safety hazard, but more about what this could truely mean, to all those nickles and dimes they've had to beg and plead for. Could cause lots of problems on actually getting true research done.
Some things I agree with and some things I don't
If you compare prices on a O3000 compared to a Starfire there prices are pretty clueful
Service/support, we have great support from SGI (of course we have lots and lots of servers). From their database side, what I had heard is that someone at SGI pissed off Larry Ellison at Oracle, so he's let SGI whither on the vine.
Poor commitment, yes when the bastard (who I shall not say) went to Microsoft was running things. I almost think he put the 320 & 520 into SGI's line just to get a job at Microsoft.
Which technology... for the 3000 we've got here I can get SCSI, FC, PCIx, Myranet, Gig E, OC3, FDDI, USB, Firewire, plus all of the special digital video stuff you want.
Longevity... no question here
Marketing has been a thorn in SGI's side ever since I've used them. Their Origin line was a kick ass server at the time, but everybody only thought of them as makers of movie effects (and still do to an extent)
The problem becomes from a price/performance metric as to what database do you use? The DB2 software itself was more expensive then the servers or the storage. 1/3 of the price of the entire system (300k+) just for the database software tends to eat into the price quickly, but I don't know of an alternative. When we checked out Oracle for Linux it was almost $200k for a single Compaq 2x PIII box, about the only option they had were MySQL or Postgress to reduce costs which would not have been up to the task.
Normally you don't want to make your filesystems that big anyway, from a backup/restore issue it becomes a killer. Any DBA worth his salt uses raw filesystems if possible, meaning you can only have one backup process on that partition i.e. normally meaning only one tape drive spinning for the entire partition. Retrieving from a single tape we can get 10-15 gig/hour, now for a restore of a 2 terabyte partition you'd be waiting a LONG while, before you could even think of applying your redo logs.
They have given an amazing level of commitment to IA32, but they have given even more to IA64 which will be very interesting IF they can pull it off. IA64 Linux numa clusters scalling to 1024 proc as a single image... that's what I call sweet.
I can deffinetly say that I like what SGI has done with the O3k series, it's cheaper than the O2K, it's a bit faster with the same mem & cpu (the faster interconnects give the boost), and I can just add an additional "p-brick" if I need more processing power without having to purchase all of the chassis stuff (XIO, disk spaces, etc). Not to mention the system partitioning capability, which I can't believe SGI slip through there fingers by allowing Sun to purchase the tech from Cray when they bought them... but that was when the Microsoft bastard was at the helm.
And a loom only *this* big, operated by swarms of bluecollar crickets. Working perfectly until they go on strike, wanting 4 weeks of paid vacation and free nectar services; throwing the production schedule of the elevator months behind schedule and overbudget.
I would say that trademarks are meant exactly for the purpose you state in the last paragraph.
Just because it is opensource doesn't make it any different than the rest of the world. I can't make a company "Tosheba" making a product named "50HD70" looking almost exactly like Toshiba's 50HX70, and expect not to have problems. If they were products who DID NOT compete no problem; but the way you put it, you are riding the coat-tails of the commercial product introducing ambiguity. Why should you want to name a competeing product the same as the commercial, except for to get the name recognition from the commercial product, which again is what trademarks are for.
If the opensource product has merits it should be able to stand on it's own, and as you say "compete" with the commercial product; hence it should not even have to have a name close to the commercial products. Look Gimp & Photoshop, the names aren't anywhere near the same, but they tend to compete against each other, and Gimp did it on it's own without having to bootstrap itself up from the name recognition of someone elses trademark.
Spelling & grammar checker off because I don't care
XFS on my beloved SGI at home does
Max Filesystem size: 18 million TB
Max File size: 9 million TB
That's according to their spec sheet, I could only dream I had 18 exabytes, course then I might need something bigger than an Indigo II, to get good use out of it.
I asked this of our rep a few months ago, asking about the port work of Linux to Mips. What he said was that Irix has a buch of special magic going on in conjunction with the Mips CPU. i.e. make sure code gets to the cpu through at the same time, special read ahead things, tweaks with the bus and cpu, etc. to put them into a Linux system would fundamentaly change how things are currently done in most of the kernel . Think how hard they've been trying just to get big memory support into the kernel, could you imaging trying to get a completely different core change that would only help Mips cpu's.
They are doing porting work, and Linux is running on Mips currently (so is NetBSD), but the special performance enhancements aren't in the kernel. Currently running Irix on a Mips Origin will be much faster than with Linux, but on the smaller workstations O2, Indigo II, etc. Linux will be faster (for non gui work) because it's so light-weight.
Don't plan on getting a speedy Linux on your Onyx, Origin systems anytime soon, unless you want to take a big performance cut.
The key word with the law is *ALLOWS*. Not being familiar with the law, but can make a reasonable stab at it from your statements. It doesn't make it illegal to make a copy of the CD's, but it at the same time doesn't say that the producer has to make it easy for you to make a copy.
I guess their reaction would be no problem. If you've been keeping up with SGI, they have been doing major things with Linux. They are selling Linux systems with excellent support (not to mention 24x7 hardware & OS support was less then IBM or Compaq by thousands of dollars). With SGI licensing their tech to nvidia if they don't use a SGI system, they are probably using a nvidia card and get a cut their. The nice thing is when someone goes, your Irix workstations are too expensive but we want your O2k server, SGI just says why don't you look at our Linux workstation with a tweaked up nvidia card that smokes everyone else.
Later this year when IA64 is hopefully released they'll roll out their Intel 3400 series line, which can scale to 512 proc's running Linux without clustering (love that numa). They currently have all of the IA32 bit offerings requested 1-4CPU small form factor servers. Once they get the IA64 goin SGI will be able to offer Linux workstations to 1U clustered servers (which they are selling today for beowulf), all they way up to a monster 512 proc numa cluster (talking with our SGI rep yesterday, someone over in the Netherlands purchased 8x 128 proc mips systems, that they then cluster to make a 1024 proc system with failover between nodes).
I feelings on the apache project things, is that they've tried unsuccessfully for so long to get the changes into the standard build, that someone said we've got to: get some traction and get it into the build, fork it into our own distro, or concede and reallocate the expensive people we've got working on it. I think SGI started to get the feeling that they were beating a dead horse.
Having just gone through a fairly rigorous review of the possiblity of using Linux for our production environment, a couple of things I've found.
NFS works, but can cause lots of grief (file handle cache, etc.)
Their are apps that cause problems with ReiserFS (Qmail has to be patched)
Root filesystem is pretty iffy
None of the journaled filesystems support nfs locking, also using a journaled fs on the root partition is pretty much use at your own risk for all of the fournaled fs at this time. Qmail has some issues with synchronous tasks with ReiserFS (FAQ at Namesys explains this)
People are throwing up ReiserFS in front of me, when I say production Linux with lots of storage of storage (we deal in lots of terabytes here...) is a "no" right now. I just throw back that if I have to worry about the app not working with my filesystem and by the way don't do any writes over NFS, then I don't feel comfortable running it in production (not to mention that nasty root compromise with ReiserFS just a few weeks ago).
Don't get me wrong I'm running both XFS & ReiserFS currently and like both, but they ain't getting anywhere near my production EMC storage, until they grow up a bit more. (on a side note we are using Linux in production, but only on things with 50 gig of storage or less, and can sustain some downtime, with limited revenue loss)
Your model assumes that the backup would be able to restore to the exact moment people were working at. We have over a hundred developers here, and if something bad were to happen to our devel environment (admins had a pc-nfs mount somewhere bad), if we where to throw out half the developers (for as you say breaks, etc) and luckily be out only 4 hours from the last backup... (which actually tends to be 8 hours for most places, others do it hourly, or every other hour)
50 workers * 4 hours = 200 lost man hours just for the developers to replace what they did since last backup + however long developers have to sit on their thumbs while the restore occurs + all the rest of the things (time diverted from other projects, etc.) you end up with a lot of time.
Your model mimics the old joke email 260 million people in the US, 114mil retired, 93 mil in school... 206,000 in hospitals... leaving only one person doing all the work.
Being in the shadow of the biggest concentration of unions (Detroit's "big three" auto industry), I would say you are completely wrong. A union is kinda like Communism, it looks great on paper but it never seems to work out like it did on the big chief notebook.
Case in point, in Ford one cannot move a monitor on ones own desk (move it over a foot to adjust for glare, etc) you have to have one of the "union" guys do it. If you do move your monitor, you get written up and the company pays the union worker for the "work" that you did... just to justify that some dumbass actually requires a job. The tech guys are not allowed to put the $50k cad workstation they built on an engineers desk they have to get a "union electrician" to plug the cables in. How completely idiotic is this? They build it, installed it, but they can't plug it into a 110 volt outlet in the wall. What they would actually do is, tell the union guy how much the workstation cost (do you want to be responsible for it, scares most union guys), and if he would let them do the work, the union guy could take credit and the pay for it. Useless absolutely useless, all from the outgrowth of unions, and the belief that because you hired a dumbass, you should never get rid of dumbass screwoffs. (Don't even get me started on the union seniority bullshit, how old you are, not your performance determines how much you get paid... ugh).
Spelling and grammar checker off because I don't care
The problem with that (and it's a very well known problem) is that it doesn't represent the country, it represents specific portions of the country; and if you think about it, it's symbolic of special interest groups.
The electorial college gives the states more power, so as to prevent candidates from just campaining in the 10 or so most populous states and essentially concede their every whim to them and screw the rest of the states of the country. Having the electorial college, actually gives the candidates reason to try and do what's best for the entire country, not just certain portions of it. This way the small guy's voice actually is paid attention to, how would you like it if you lived in somewhere less populous and because the vote was based entirely on the popular decision the candidates went to Cal & NY and said everyone's taxes in Montana, Wyoming, Kansas, etc. will go to paying Cal & NY taxes so they won't have to pay any this year because they are going to vote for me... now think about not being able to do a damn thing about it because the 10% of states they're giving tax breaks to have more votes than the rest of the 90% combined.
The candidates are responsible for the nation as a whole not to just kiss the ass of who's got the most people living in certain portions.
Spelling and grammar checker off because I don't care
The previous poster didn't state it very well, but his general "gist"(sp?) was correct.
To use your analogy, being attacked with a hammer is not categorized as assault, but assault with a deadly weapon; the hammer is not illegal to use, but to use it for assault, etc. is. That's what this treaty is trying to do, you use a tool in a manner that is against the law, your mis-use of the tool can be used against you.