This is amazingly petty of me, but I can't stomach using OpenOffice since they changed the name of the software to "OpenOffice.org". What, was "OpenOffice" not getting the point across?
Sigh. I suppose I should at least be thankful that they didn't call the individual components calc.openoffice.org, writer.openoffice.org, and so on.
However, pettiness aside, to the uneducated, the ".org" at the end is thoroughly confusing. Ever tried recommending it to someone? "Hey, you should switch from Microsoft Office to OpenOffice.org!" "Oh, you mean I can create my spreadsheets on the web?"
Perhaps if they spent more time making it less resource-intensive than Microsoft Office, it might actually be a contender. For now, I'll stick with Microsoft Office on my Mac when I have to use it, and LaTeX otherwise. PDFs for everybody!
and it's still not hard to find links with kazaa lite, or anything to do with illegal nature.
< snip >
or is it a case of one rule for one, another for everybody else?
This is precisely why they shouldn't filter out any search results. If they filter only some some "illegal" (keep in mind that legality is a function of locality) links, they are, in effect, endorsing the others. Your sacred cow may not be mine, and everyone's kink is someone else's horror.
Search engines merely provide a searchable index to content. I don't see how they can be seen as guilty of copyright infringement, so long as the page descriptions are kept short enough to qualify under "fair use". An impartial tool like a search engine is not a place to enforce morality, since morality is subjective by its very nature.
I should close by saying that I don't personally condone the examples you've given, but the people involved in the creating of that content are the wrongdoers, not Google.
I have two on my desk right now (an O2 and an Octane), and a couple servers in colo.
You seem to be forgetting that some people use their computers for work at work rather than playing the latest game at home. SGI systems are extremely good at what they do, and they make bad-ass systems for almost any problem that needs a lot of memory bandwidth.
But, yes, it'd be hard to justify a $40k workstation to play Unreal Tournament. It'd also be hard to justify an 18-wheeler to drive to the office every morning. It's all about situation and perspective.
However, used SGIs can be had for cheap-cheap on eBay. Try one sometime. If you keep an open mind, the SGI bug will bite you, and someday, you too might have an Onyx XL in your dining room.:)
In other news, doughnut mogul Krispy Kreme (NYSE: KKD) has placed a bid to purchase high-end computing giant Sun Microsystems (NASDAQ: SUNW) in order to secure a first-in-line place for the much anticipated ``blade'' form-factor servers with UltraSPARC IV CPUs to control production, shipping, and doughnut-glazing processes. Officials at Sun Microsystems could not be reached for comment, but a guy hauling garbage out of the rear door of building 7 of the San Diego campus was quoted as saying ``I really like them doughnuts. If they buy us, maybe we'll all get doughnuts for free at lunch. That'd rule.''.
Look out Dell and IBM, when you've got a million doughnut and coffee addicts for your clients, switching them to Unix and Java can't be far behind.
After all the money you spent on that MCSE, the MSDN subscriptions, etc..., why would you be so kind as to give out tips to an anonymous fellow on the internet?
Perhaps not in this forum, but if anyone really wants a good Microsoft tech forum, take a look-see on Microsoft's public news server: msnews.microsoft.com. Every Windows-oriented weblog that I've seen is utter crap (Including the MSDN comments forums, which are usually strewn with comments like "D000D! I d0wnl0aded the lat3st DirectX SDK, and why aren't my games any faster?").
The really ironic (or perhaps intentional?) thing is that most Windows-weenies don't even know what USENET is, yet MSCE volunteers are giving the absolute best tech advice on newsgroups sponsored by Microsoft. The signal-to-noise ratio is probably about equivalent (or better) to any random c.o.l group.
There's a dirty little secret here called a "preinstallation kit". I used to be a PC OEM, and I was a member of the Microsoft System Builder program (which is basically required to buy the OEM-discounted OS media).
Basically, to get the OEM discount, the OEM accepts a license agreement that they can only get rid of the OEM software:
Preinstalled on a PC
Uninstalled, but bundled with an unassembled computer, a motherboard, or a mass-storage device
As a "Multipack" (usually 5 OEM copies in a sealed kit)
Now, "preinstalled" doesn't mean what you think it does. It does, in fact, mean that the software resides on the computer's hard drive, but there is a very specific way it gets there. Just running the "end user" installer isn't sufficient, and will cause you to fail an audit.
However, the way that OEMs do have to install the software is completely automated. The OEM supplies a floppy with some installation options, and the little OEM bitmap that goes in the "system properties" control panel, and Microsoft supplies a preinstaller CD that automates the building of the installed image on the computer.
Let me reiterate: Except for the initial effort of creating a master preinstallation floppy for each OS, the process is no different across all the Windows family. Really, it'd be a nice system if the OEMs didn't have the proverbial gun to their heads. That attitude is a big part of the reason I quit selling PCs.
Now, if a Linux distribution, or FreeBSD could devise such an automatable installer, especially if it could do the dual-booting voodoo automagically, they'd have a case: "but it's as easy to integrate as Windows!". As it is, the default installers are still interactive, which makes for difficult bundling when you try to churn out PCs en masse.
My Computer can be renamed, though I forget exactly how
Highlighting it and renaming it, just like any other object, has always worked for me.
Recycle Bin is the only thing I haven't figured out how to permanently rename/get rid of.
Edit the "default" value in the registry key HKEY_CLASSES_ROOT\CLSID\{645FF040-5081-101B-9F08-0 0AA002F954E}. Mine's been "Bit Bucket" since Win95. If you pay careful attention to how COM objects in the registry work, and how they interact with the shell, you can probably figure out how to conjure up and destroy these "special" folders and desktop objects at will[1]. It's a really neat (if underexposed) system.
Note: Slashcode has probably broken that key reference. There are no spaces in it[2].
I've always liked IRIX's "dumpster", complete with a gratifying th-thunk sound when you drop an offending file into it. However, it'd be really nice if it didn't just behave like a regulal folder (specifically, it'd be nice if it remembered where to restore files to, instead of just holding them).
[1] This is left as an exercise for the reader. It's documented in MSDN, if you really care. [2] Apparently[3], the "lameness filter" now introduces lameness to balance-out the lameness that it previously removed. Hmm. [3] Yes, I know the reason, and I think the cure is worse than the disease.
I can't wait to see Monster making Ethernet cables.:) Just think of it: Ethernet cables you can stomp on, twist, bend unmercifully, and they still work perfectly--and the manufacturer will replace it if it doesn't.
Maybe I've been spending way too much time tinkering with computer security, but does anyone else think that this might be a really neat (if infeasible) system of biometric authentication? Example: "To enter this facility, please swipe your card or play the first 15 bars of Stevie Ray Vaughan's Rude Mood."
If nothing else, it'd be a really neat hack. I'd love to be able to "bless" a workstation (we have roaming admin privileges here) by plugging a guitar into an adjacent port and jamming, rather than slipping in my smart card.
My company had a similar experience with UPS. We received two Sun SPARCarray 100-series disc chassis from New Jersey (shipped to Texas).
Now, if you're not familiar with these particular pieces of Sun kit, they're 75-pound rackmountable steel safety cages with some SCSI and FCAL intelligence. If you drop one on your foot, you will be laid up for a week or two with a broken foot.
In short, they're just the sort of thing that OSHA doesn't want your employees to be lugging about, but they're exactly what you'd trust to keep your discs safe from accidental bumps and vibration.
We received one double-boxed with foam-rubber and styrofoam padding. The first words out of my floor tech's mouth was "Uhm, it's round." The box was damn-near spherical, and the array itself was twisted. We received the other one, originally boxed the same way, but afterwards in a completely trashed (shredded?) box. Both were absolutely worthless.
Really, unless UPS typically ships via trebuchet, I have no way of knowing how this happened.
More recently, I received an SGI Octane shipped via UPS (I forgot to tell the vendor to ship FedEx). Nothing in that machine was straight. Again, if you've never seen an Octane in-person, you need to know that it takes two people to safely move it any distance--it's a very sturdy computer. The partitions between the XIO bays were rippled, the computer wouldn't POST, the skins were crunched, and all the access handles were broken-off.
Luckily, since money had changed hands across both transactions, it was fairly easy to tell UPS exactly what they owed us. We got replacement equipment each time, and the vendor was reimbursed.
Since you're shipping your own stuff, UPS will probably try to strong-arm you into shutting up. Don't. Get an attorney. If nothing else, get him/her to write a letter on your behalf. Get an appraisal of what the equipment was worth, and get an appraisal of what the equipment would cost to repair. The most you can expect to get out of them is the lesser of the two appraisals plus legal fees.
As other people have told you, this is a great (albeit painful) opportunity for a lot of people to learn a lesson: If the box looks mangled, do not sign for it. Ever. Tell UPS/FedEx/whomever to take it back, and tell them that you want a claim form on the spot. This will immediately trigger an investigation, and the package will be taken back to the carrier's inspection center.
Odd spell-checker you have there. Here's a test for you: Design a system of objects that contain at least two independent characteristics (say, for example, in a MOO, an object can either be enterable (like a room or a bag), and carryable (like a sword)). Now, implement it in such a way that a class can have any subset of those characteristics.
Hmm... that'd be pretty easy with multiple-inheritance. Since you're doing the exact same thing on each object, you could just inherit an object that implements that characteristic. However, you can't do that in Java, so you're stuck implementing (key word here) the interface on each class (since, remember, any class may have any subset of characteristics, so you can't rely on direct inheritance). Thus, redundant code.
Even if you use the visitor pattern, you're still reimplementing the visitor boilerplate in each class, instead of just an inheritance clause.
I suppose you could implement each feature as a bean, and the define each dynamically, but your constructors would all look differently, and it would be a bitch to fix if you ever needed to change the bean-interface.
Try creating a memory leak or mishandle a pointer in Java. What? You cant do that in Java?
Try using Swing and Java3D (or any other JNI-based library) in the same application. Then, try loading a huge (at least 128 megabyte) dataset into the application in such a way that the data need travel over the JNI link.
That's right, using only Sun-supplied code, you too can completely bomb your application in portable ways under Windows NT, IRIX, and Solaris! It seems that Java GCs large objects in such a way that pointers drift while the JNI-linked libraries need them to be fixed.
Java is not a panacea. I'd say that something like Qt (combined with the STL) is much closer to what the original poster was looking for. It's portable, it's C++, which is what the poster was asking for, and it's far more scalable for a desktop app.
Heh... I'd completely forgotten about that. We've got a fair number of customers on that. It's a lot more feasible with big terminal servers, though, and that incurs a cost--you can't shtogun across multiple terminal servers, so the ISP has to eat the cost of high-end gear up front. In our case, that's not a problem, since our last purchase was a consolidation of a bank of Lucent gear.
The only problem I've seen in that Windows (read as: the OS used by > 90% of my customers) doesn't seem to want to sync the two modems up (reliably) unless the two modems are using the same driver, and the majority of WinModems (at least the Lucent ones) will only allow one instance per PCI bus.
So... you're left looking for hardware modems. I could rant for hours about how customers don't want to pay the extra $20 for a used USR Courier (or another -decent- hardware modem), and then bitch about connection quality, but I'll save that for another day.:) Suffice it to say that those people will never benefit from shotgunning, since WinModems tend to suck at it.
I agree, though. Every ISP should support it. Every modern terminal server does. Any terminal server that supports ISDN does. It's just a configuration change (turn on MPP support), and it can generate a lot of revenue (charge an extra $10/mo per port--it's still cheaper than ISDN or DSL) and a lot of happy customers, if they're willing to buy decent client-side kit.
I certainly believe stability is FAR more important than increased speed, and being on the @ss end of an ISP, I'd much rather have a reliable connection -- but there's a limit to how long you can blame the vendors.
If the vendors won't release updates in a reasonable manner (IE: *NOT* require you to purchase all new equipment), then dump them for someone who will.
And blame whom, then? One needs hardware to connect, and a decent terminal server is not cheap hardware by a long shot--you can't exactly build (a scalable) one from commodity components and free software. The last one my company purchased was a $1.2M investment (several hundred digital modems and a few dozen PRI ports). Liquidating it on eBay and blowing that much money again on a piece of equipment strictly for an incremental improvement on the customer side isn't an option--not because I don't care about my customers, but because it isn't economically feasible to do this every time J Random Vendor happens to lose a clue on upgrades.
And, yes, we got screwed on the deal. We also have to replace our modem cards (and the shelf controller!!) to put v.92 on the wire. Everyone involved (except for the vendor, waiting with open arms for our next purchase order) is pissed about the ordeal, but there's really not a damn thing we can do.
It's not like you can evaluate equipment on the basis of what standards will be ratified N years down the road, since we don't know which standards will be ratified. We can guess, but the industry has been surprised before.
If the upgrades needed aren't in your budget, than you didn't do your budget properly.
This isn't an upgrade. This is a complete solution replacement, and almost all the vendors dropped the ball on this one, so shut your pie hole.
Hardware shouldn't be considered to have a lifespan much beyond 2 years nowadays.
Bullshit. Absolute, complete, and total grade-A bullshit. This is the first upgrade in at least five years that has completely broken the server-side of the ISP. V.90 was a firmware update for every major vendor (except for 3com, which never gets upgrades right). The per-modem cost of a decent terminal server is astronomical, and no company can recover that cost in two years and maintain a decent customer-to-modem ratio.
Now, maybe your PC doesn't have a lifespan beyond 2 years, but networking hardware Just Isn't That Way.
I suggest you stay on the @ss end of your ISP for now. You don't appear to understand what is actually going on.
The real reason most ISPs do not support v.92 right now is bacause there are no reliable server-side implementations. 3com has an update package consisting of "replace everything at 80% of the cost you just invested last year". Alcatel says "replace everything, oh, and, by the way, we changed all the part numbers again, so your catalogs are out of date for the third time this year, also, since this is new and better equipment, be prepared to pay 20% more than you did last year". ciscoSystems says "here, this might crash your terminal server and void your TAC onsite agreement, but, then again, it might work flawlessly--do ya feel lucky, punk?". Lucent is still trying to iron out the bugs in its v.90 implementation, and they've been contemplating EOLing the TNT modem cards for years now--they've only held off, as they don't have a replacement part that works yet.
Believe me, there are far more factors in this equation than "$ISP doesn't want to fork out $20 and a few minutes to upload some firmware". Most of us just have our hands tied behind our backs because we just bought a buttload of equipment a year or so ago, and, with the economy being what it is, don't have the funds to replace all of it.
FWIW, I connect via dialup at home. I can't say that I'm waiting for v.92 with bated breath. v.90 is fine for me, and, from what I've seen, v.92 is only incremental--barely noticeable except in absolutely perfect conditions. Basically, if your phone line and ISP are such that you can achieve the highest-quality v.92 connection between the two points, you can probably get DSL, due to the stringent line quality standards for v.92. With that in mind, a lot of companies (including the hardware vendors) are pushing DSL a lot harder than v.92 for customers that want better upload speeds.
Read the EULA. You can purchase Windows 2000 seats through volume-licensing and "downgrade" the installation to NT 4. This applies to both the server and workstation editions.
Then, once you and your staff are trained on Windows 2000, you can upgrade, as you've already purchased the right (for n seats, anyway).
I don't know if this applies to shrinkwrap licenses, but that's how it is with volume licensing. Also, I think you can still get NT 4 media kits through the "easy fulfillment" program, if you're a volume-license customer.
I'm not carrying the Microsoft torch here or anything, but your statement is patently false. It is, however, what Microsoft wants you to believe, which is why they insinuate it, and only contradict it in the fine print.
I'd like to address the "just view the damned ads, you freeloading hippies" crowd.
Personally, the reason I started blocking banner ads (a little over a year ago) was because of one very specific ad--that stupid "punch the monkey ad".
It managed to crap more web no-nos into an ad than I ever though possible:
I froze my browser, as my browser had to load the Java runtime to display it. This is nontrivial time under Netscape, and used to be a lengthy wait under IE, as well.
It moved. Quickly. Very distracting when you're trying to use Altavista to look up a particular bit of LaTeX wizardry.
If my mouse cursor hovered over the ad, the ad captured mouse focus, and caused my mouse cursor to not always move as it normally would (largely due to the overhead by the Java runtime, I'm sure--I was using a SPARC LX at the time).
It would frequently cause Netscape to dump core, and would occasionally cause IE to just freeze-up completely.
At the time, it was a very popular ad. I don't know what I was typing to into Altavista to make it trigger (LaTeX->latex? Monkeys? WTF?), but I seemed to get it every five pages, and Netscape dumping core every five pages was not conducive to my finding out this LaTeX technique, which I needed right then to finish a CS paper (I'd have used Fondren Library, but this was before the Rice campus library stayed open 24 hours daily).
So, as a temporary fix, I disabled Java (I didn't need it at the time), used a different search engine (Google), got what I needed, and then installed Squid+Cameron Simpson'sAd Zapper (once I'd turned-in my paper), and the problem went away. I could have Java as I needed it (Rice's CS departmnet loves Java. Turning it off in a web browser meant not being able to do certain coursework), and my browser didn't crash because of stupid monkeys.
The clear message I'd like to deliver is I don't mind non-intrusive advertising. In fact, most banner ads are very interesting, so long as they don't flash or titter about annoyingly, and don't stupidly try (and fail) to look like dialog boxes (looks really stupid under OpenWin). Occasionally, I click one. However, if it pops up in a separate window, if it spawns things in other windows, if it creates offscreen windows, if it crashed my browser, if it litters my hard drives with cookies, if it prevents me from clicking on your page, or if it dances around like a stupid monkey, I will disable it, and I will go elsewhere.
There are probably a lot of technically-minded users that feel the same way. I don't want to steal content--I don't have this need to remove all adverts from the pages I'm viewing (although, I will strip them out, if need to print the page). But, my computer is my computer, and if your website can't sit in its window and behave itself, you've just lost a viewer.
That should work great, provided you've got the RAM (more than 128M, on 6.5) and the CPU (R10k, or maybe R4400/250). Max Impact+TRAM is just an older version of MXI, without the XIO cross-connect between the CPU and gfx (which is limited to the Octane line, anyway). Basically, if SGI still published the periodic tables, you'd see MXI and Max Impact+TRAM on the same row--it's the same hardware on a different bus.
I wish I had one, but I'm getting an Octane/SI (to replace my trusty Indigo2 Killer Impact) on Monday!:) I don't do enough 3D to justify the texture memory, anyway--I just use it as a devel station because IRIX is a Kick Ass Unix(TM).
I've tried GLQuake on an Indigo2 XZ before, and it was pathetic due to the lack of hardware texturing--it handled the geometry just fine.
Now -this- is the stuff of nightmares.
on
HP Buys Compaq
·
· Score: 2, Informative
Compaq doesn't just make PCs, and HP doesn't just make PCs. They both make "high-end" (read as: expensive and unreliable) servers and workstations. What if they merge the "high-end" divisions? Can we look forward to:
OpenVMS on PA-RISC (not too scary, but... ewww), now that the Alpha is gone?
HP-UX on Alpha, just to help Intel rape the last bit of dignity out of the platform?
A resurgance of NT/MIPS with the new "Kayak Himalaya" workstations (dumb, but this is Compaq and HP we're talking about here)?
PA-RISC in a new line of "Nonstop Himalaya" server (again, dumb, but is MIPS over Alpha was stupid, too)?
and, the scariest, and most-probable....
An HP-made server, with that <sarcasm>wonderful</sarcasm> HP support, running HP-UX (blech) on IA64 (cringe), using Compaq's horribly nonstandard system components and chassis, with HP's horribly nonstandard (and flaky) RAID system.
I just hope HP sells of the stuff they make that doesn't suck (calculators, printers, and medical/testing equipment) before they make something really stupid and tank. Or, maybe we'll just luck-out and simply continue to make the same crappy PCs--all under one roof.
I think I need some liquid recovery now.
Before we had an upstream that provided news to us, we used a "suck" server as a solution.
Basically, you run an NNTP server that acts just like a web-cache: it fetches the headers, and caches them. Then, when you download an article, it caches that, too. It also checks the upstream server for expiration and generally mimics the upstream server.
The only real downside (for the customers) is that all groups initially appear empty, except for a post that says something to the tune of "articles are downloading--please come back in a few minutes".
It's a really simple solution--your customers get news, and you don't consume a full newsfeed. If you know someone with clout at your ISP, you might want to point out that solutions like this exist (and, IIRC, are "Free").
That's a staggeringly good idea. Maybe we could even get our words and culture back. It's even case-worthy (IMHO, IANAL), since we can't admit that we're members of that culture without a negative response from Joe Average (think of the racial discrimination parallels).
The big difference that I see is that hackers are not members of any particular country, and the relavent bit of the Hague convention is designed to make IP laws enforceable under the rules of the owning party's locale.
But, at least American hackers would have a good chance--the US seems to love IP-law enforcement.
I've usually had to include rm ~/core, as well, and call it from my.login file, as Netscape had a bad habit of crashing X on Linux and IRIX 5.
I think the really odd thing is that Netscape is very stable on the Macintosh. It sucked on IRIX, Solaris, Linux, Win32, Win16 (what didn't?), and Digital UNIX, but it really ran well on MacOS--better than your average Mac application.
....anyway.... I think the "What happens to Netscape" was intended to mean "Does this mean that AOL is no-longer supporting Netscape?". What about AOL/Macintosh, does it demand IE, as well? I know a lot of Mac users who won't even install Office because it's a Microsoft product--you think they'd install the program that they're convinced "killed the only decent web browser by playing dirty"?
Then again, there's really no reason they can't support Netscape 4 or Mozilla on platforms that don't yet have IE. Maybe this is just overblown like they typical Slashdot fare. Linux Is Not UniX
I somehow get the feeling that this is not a totaly unbiased opinion. Below are answers to some of you problems.
Well, it's my personal experience, so any bias that shows through is as a result of me actually using both systems for years on SPARC hardware. To the best of my ability I've given both operating systems an equal chance, and I still have Linux, IRIX, and Solaris hosts on my network. I'm really only biased towards the IRIX boxen, though, which (thankfully) haven't entered into this discussion yet.:)
If your suspicions are because of my.sig, it's there just because I'm sick of interviewing job applicants that supposedly have "loads of Unix experience", yet sit down at an IRIX or Solaris console and cry "where's GNOME? Why isn't my shell bash? Where do I get the RPM for it? Does this version of IRIX have the new 2.4 kernel? Why can't I #include anything in <linux/>?". So, a sign bearing that sentence (with demonstrated emphasis) is a permanent fixture on my desk.
The vendors who tell me to "reboot into Windows and see if the problem is still there" don't help the morale, either, but that's another rant for another day....
You are right. In order to get good system recovery, why not install sash? apt-get install sash, and all is well.
I'm sure I could download the source to everything necessary for recovery, and build it statically (in fact, I have done this on my PC running Slackware), or download n packages to do it for me. The point is that this is part of a standard Solaris installation./sbin/sh is static./sbin/* are static.
I'm comparing the environments by what I get when I do a fairly reasonable install of each on the same platform, not by what I can download or hack-together--if we were to use that as a basis, all operating environments would be approximately equal.
Have you tried e2label - I haven't used linux on sparcs, but on my intel machine, you can label your hard disks, to get consistant mounts.
I haven't heard of it, but (from reading the manual page) it seems to work at the fileystem level. How can that possibly affect the order in which the kernel detects devices and assigns them to node-numbers?
I wasn't saying that my partitions were drifting, my devices showed up in all sorts of odd orders, and they shift when you add or remove one. Solaris labels discs by controller number, SCSI ID, and LUN, so if you know that data, you know the name of the device, and attaching labels to anything is only a convenience--not a necessity for a consistent system.
Yes, emulation is slow, and incomplete. There are some SunOS system calls that are unimplemented, beacuse they require hardware features that Linux does not yet suppport. All of this really wouldn't matter if there were commercial apps for S/Linux.
I honestly did not know that LD_LIBRARY_PATH worked in Linux. I stand corrected. Thanks for pointing this out!
The buzzword standard is POSIX, now known as Unix95 and Unix98. That's the Unix standard.
To the best that I can tell, the free software untiites conform to the "I thought it would be cool to..." standard. Sort of like "I thought it would be cool to dump man pages and use info pages, even though man exists on every Unix machine on the planet, and I have to get info from ftp.gnu.org and it requires GNU terminfo and GNU readline and...". In some circles, GNU stands for "Get New Utilities".
If you know Unix, as a user, you know Solaris as a user. I came from an OSF/1 (ow) and IRIX background, and hit the ground running (because of the POSIX standards). I used Solaris and IRIX as a user and a developer long before I was had to manage the systems, and this adherance to standards made life easy both as a user and a sysadmin.
It's not an obvious case where free software wins anything, it's a difference in paradigms: do you add every imaginable bell and whistle, or do you stick by the rules and do what people expect. For some people, one paradigm works, for some the other works. I happen to like consistency--I think it makes life easier.
I'm not sure on the 64-meg minimum (in the license?), as I've done test installs on machines with as little as 32 megabytes (the minimum to load suninstall).
You must just be having bad luck with hard drives. If probe-scsi in OpenBoot can see it, then Solaris will see it, since Solaris just copies the OpenBoot device tree to/devices. I've got a mix of Sun, SGI, generic, and Compaq drives in one of my SPARCs, and Solaris 8 (obtained under the FBL) installs without problems.
Are you sure the problem isn't that you forgot to make a Sun disklabel on the drive? If the installer boots screaming about bad magic numbers and misking disklabels, you forgot to run format on your disc. When suninstall loads (I'm assuming you're not doing the webstart install), enlarge the console window, and type format. Tell it to label every disc--no need to partition, suninstall can do that automagically.
If you don't label the disc, the result will be that the installer ignores it, assuming that it's in use by a non-SunOS operating system.
Of course it's not documented, because if you bought disc drives from the evil marketroids at Sun, they'd be prelabeled, and you'd never have to fiddle with format.
Then again, you have to run fdisk manually in Slackware, too, so I suppose it's not horribly crufty.
Oh, poo. You're just ticked off because Sun switched from a BSD-style system to a SYSV-style system in SunOS 5 and then retroactively renamed SunOS 4 to Solaris 1, thinking that their customers would never notice the difference.
I feel your pain. SunOS 4.1.4 is probably my favorite OS on the planet. It's not Unix if you don't have to lay out your discs in cylinders, right?:)
Seriously, though, SunOS > 5.4 is really nice. It's finally as stable as the OS they had for years before. It's still not BSD, but it's better than a lot of other SYSV systems out there, and you can still run your old BSD-style apps, too, since they finally got SunOS 4 emulation working.
Slashdot Mirror
I brake for tailgaters, and then I sue. Or, occasionally, I'll donate a nickel out the T-tops to the "buy-the-driver-a-clue" foundation.
This is amazingly petty of me, but I can't stomach using OpenOffice since they changed the name of the software to "OpenOffice.org". What, was "OpenOffice" not getting the point across?
Sigh. I suppose I should at least be thankful that they didn't call the individual components calc.openoffice.org, writer.openoffice.org, and so on.
However, pettiness aside, to the uneducated, the ".org" at the end is thoroughly confusing. Ever tried recommending it to someone? "Hey, you should switch from Microsoft Office to OpenOffice.org!" "Oh, you mean I can create my spreadsheets on the web?"
Perhaps if they spent more time making it less resource-intensive than Microsoft Office, it might actually be a contender. For now, I'll stick with Microsoft Office on my Mac when I have to use it, and LaTeX otherwise. PDFs for everybody!
If $company has servers or offices in that locality, by all means! But, aren't all of Google's servers here in the US?
This is precisely why they shouldn't filter out any search results. If they filter only some some "illegal" (keep in mind that legality is a function of locality) links, they are, in effect, endorsing the others. Your sacred cow may not be mine, and everyone's kink is someone else's horror.
Search engines merely provide a searchable index to content. I don't see how they can be seen as guilty of copyright infringement, so long as the page descriptions are kept short enough to qualify under "fair use". An impartial tool like a search engine is not a place to enforce morality, since morality is subjective by its very nature.
I should close by saying that I don't personally condone the examples you've given, but the people involved in the creating of that content are the wrongdoers, not Google.
I have two on my desk right now (an O2 and an Octane), and a couple servers in colo.
You seem to be forgetting that some people use their computers for work at work rather than playing the latest game at home. SGI systems are extremely good at what they do, and they make bad-ass systems for almost any problem that needs a lot of memory bandwidth.
But, yes, it'd be hard to justify a $40k workstation to play Unreal Tournament. It'd also be hard to justify an 18-wheeler to drive to the office every morning. It's all about situation and perspective.
However, used SGIs can be had for cheap-cheap on eBay. Try one sometime. If you keep an open mind, the SGI bug will bite you, and someday, you too might have an Onyx XL in your dining room. :)
In other news, doughnut mogul Krispy Kreme (NYSE: KKD) has placed a bid to purchase high-end computing giant Sun Microsystems (NASDAQ: SUNW) in order to secure a first-in-line place for the much anticipated ``blade'' form-factor servers with UltraSPARC IV CPUs to control production, shipping, and doughnut-glazing processes. Officials at Sun Microsystems could not be reached for comment, but a guy hauling garbage out of the rear door of building 7 of the San Diego campus was quoted as saying ``I really like them doughnuts. If they buy us, maybe we'll all get doughnuts for free at lunch. That'd rule.''.
Look out Dell and IBM, when you've got a million doughnut and coffee addicts for your clients, switching them to Unix and Java can't be far behind.
Perhaps not in this forum, but if anyone really wants a good Microsoft tech forum, take a look-see on Microsoft's public news server: msnews.microsoft.com. Every Windows-oriented weblog that I've seen is utter crap (Including the MSDN comments forums, which are usually strewn with comments like "D000D! I d0wnl0aded the lat3st DirectX SDK, and why aren't my games any faster?").
The really ironic (or perhaps intentional?) thing is that most Windows-weenies don't even know what USENET is, yet MSCE volunteers are giving the absolute best tech advice on newsgroups sponsored by Microsoft. The signal-to-noise ratio is probably about equivalent (or better) to any random c.o.l group.
- Preinstalled on a PC
- Uninstalled, but bundled with an unassembled computer, a motherboard, or a mass-storage device
- As a "Multipack" (usually 5 OEM copies in a sealed kit)
Now, "preinstalled" doesn't mean what you think it does. It does, in fact, mean that the software resides on the computer's hard drive, but there is a very specific way it gets there. Just running the "end user" installer isn't sufficient, and will cause you to fail an audit. However, the way that OEMs do have to install the software is completely automated. The OEM supplies a floppy with some installation options, and the little OEM bitmap that goes in the "system properties" control panel, and Microsoft supplies a preinstaller CD that automates the building of the installed image on the computer. Let me reiterate: Except for the initial effort of creating a master preinstallation floppy for each OS, the process is no different across all the Windows family. Really, it'd be a nice system if the OEMs didn't have the proverbial gun to their heads. That attitude is a big part of the reason I quit selling PCs. Now, if a Linux distribution, or FreeBSD could devise such an automatable installer, especially if it could do the dual-booting voodoo automagically, they'd have a case: "but it's as easy to integrate as Windows!". As it is, the default installers are still interactive, which makes for difficult bundling when you try to churn out PCs en masse.Highlighting it and renaming it, just like any other object, has always worked for me.
Edit the "default" value in the registry key HKEY_CLASSES_ROOT\CLSID\{645FF040-5081-101B-9F08-0 0AA002F954E}. Mine's been "Bit Bucket" since Win95. If you pay careful attention to how COM objects in the registry work, and how they interact with the shell, you can probably figure out how to conjure up and destroy these "special" folders and desktop objects at will[1]. It's a really neat (if underexposed) system.
Note: Slashcode has probably broken that key reference. There are no spaces in it[2].
I've always liked IRIX's "dumpster", complete with a gratifying th-thunk sound when you drop an offending file into it. However, it'd be really nice if it didn't just behave like a regulal folder (specifically, it'd be nice if it remembered where to restore files to, instead of just holding them).
[1] This is left as an exercise for the reader. It's documented in MSDN, if you really care.
[2] Apparently[3], the "lameness filter" now introduces lameness to balance-out the lameness that it previously removed. Hmm.
[3] Yes, I know the reason, and I think the cure is worse than the disease.
I can't wait to see Monster making Ethernet cables. :) Just think of it: Ethernet cables you can stomp on, twist, bend unmercifully, and they still work perfectly--and the manufacturer will replace it if it doesn't.
Maybe I've been spending way too much time tinkering with computer security, but does anyone else think that this might be a really neat (if infeasible) system of biometric authentication? Example: "To enter this facility, please swipe your card or play the first 15 bars of Stevie Ray Vaughan's Rude Mood."
If nothing else, it'd be a really neat hack. I'd love to be able to "bless" a workstation (we have roaming admin privileges here) by plugging a guitar into an adjacent port and jamming, rather than slipping in my smart card.
My company had a similar experience with UPS. We received two Sun SPARCarray 100-series disc chassis from New Jersey (shipped to Texas). Now, if you're not familiar with these particular pieces of Sun kit, they're 75-pound rackmountable steel safety cages with some SCSI and FCAL intelligence. If you drop one on your foot, you will be laid up for a week or two with a broken foot. In short, they're just the sort of thing that OSHA doesn't want your employees to be lugging about, but they're exactly what you'd trust to keep your discs safe from accidental bumps and vibration. We received one double-boxed with foam-rubber and styrofoam padding. The first words out of my floor tech's mouth was "Uhm, it's round." The box was damn-near spherical, and the array itself was twisted. We received the other one, originally boxed the same way, but afterwards in a completely trashed (shredded?) box. Both were absolutely worthless. Really, unless UPS typically ships via trebuchet, I have no way of knowing how this happened. More recently, I received an SGI Octane shipped via UPS (I forgot to tell the vendor to ship FedEx). Nothing in that machine was straight. Again, if you've never seen an Octane in-person, you need to know that it takes two people to safely move it any distance--it's a very sturdy computer. The partitions between the XIO bays were rippled, the computer wouldn't POST, the skins were crunched, and all the access handles were broken-off. Luckily, since money had changed hands across both transactions, it was fairly easy to tell UPS exactly what they owed us. We got replacement equipment each time, and the vendor was reimbursed. Since you're shipping your own stuff, UPS will probably try to strong-arm you into shutting up. Don't. Get an attorney. If nothing else, get him/her to write a letter on your behalf. Get an appraisal of what the equipment was worth, and get an appraisal of what the equipment would cost to repair. The most you can expect to get out of them is the lesser of the two appraisals plus legal fees. As other people have told you, this is a great (albeit painful) opportunity for a lot of people to learn a lesson: If the box looks mangled, do not sign for it. Ever. Tell UPS/FedEx/whomever to take it back, and tell them that you want a claim form on the spot. This will immediately trigger an investigation, and the package will be taken back to the carrier's inspection center.
Heh... I'd completely forgotten about that. We've got a fair number of customers on that. It's a lot more feasible with big terminal servers, though, and that incurs a cost--you can't shtogun across multiple terminal servers, so the ISP has to eat the cost of high-end gear up front. In our case, that's not a problem, since our last purchase was a consolidation of a bank of Lucent gear.
The only problem I've seen in that Windows (read as: the OS used by > 90% of my customers) doesn't seem to want to sync the two modems up (reliably) unless the two modems are using the same driver, and the majority of WinModems (at least the Lucent ones) will only allow one instance per PCI bus.
So... you're left looking for hardware modems. I could rant for hours about how customers don't want to pay the extra $20 for a used USR Courier (or another -decent- hardware modem), and then bitch about connection quality, but I'll save that for another day. :) Suffice it to say that those people will never benefit from shotgunning, since WinModems tend to suck at it.
I agree, though. Every ISP should support it. Every modern terminal server does. Any terminal server that supports ISDN does. It's just a configuration change (turn on MPP support), and it can generate a lot of revenue (charge an extra $10/mo per port--it's still cheaper than ISDN or DSL) and a lot of happy customers, if they're willing to buy decent client-side kit.
And blame whom, then? One needs hardware to connect, and a decent terminal server is not cheap hardware by a long shot--you can't exactly build (a scalable) one from commodity components and free software. The last one my company purchased was a $1.2M investment (several hundred digital modems and a few dozen PRI ports). Liquidating it on eBay and blowing that much money again on a piece of equipment strictly for an incremental improvement on the customer side isn't an option--not because I don't care about my customers, but because it isn't economically feasible to do this every time J Random Vendor happens to lose a clue on upgrades.
And, yes, we got screwed on the deal. We also have to replace our modem cards (and the shelf controller!!) to put v.92 on the wire. Everyone involved (except for the vendor, waiting with open arms for our next purchase order) is pissed about the ordeal, but there's really not a damn thing we can do.
It's not like you can evaluate equipment on the basis of what standards will be ratified N years down the road, since we don't know which standards will be ratified. We can guess, but the industry has been surprised before.
This isn't an upgrade. This is a complete solution replacement, and almost all the vendors dropped the ball on this one, so shut your pie hole.
Bullshit. Absolute, complete, and total grade-A bullshit. This is the first upgrade in at least five years that has completely broken the server-side of the ISP. V.90 was a firmware update for every major vendor (except for 3com, which never gets upgrades right). The per-modem cost of a decent terminal server is astronomical, and no company can recover that cost in two years and maintain a decent customer-to-modem ratio.
Now, maybe your PC doesn't have a lifespan beyond 2 years, but networking hardware Just Isn't That Way.
I suggest you stay on the @ss end of your ISP for now. You don't appear to understand what is actually going on.
The real reason most ISPs do not support v.92 right now is bacause there are no reliable server-side implementations. 3com has an update package consisting of "replace everything at 80% of the cost you just invested last year". Alcatel says "replace everything, oh, and, by the way, we changed all the part numbers again, so your catalogs are out of date for the third time this year, also, since this is new and better equipment, be prepared to pay 20% more than you did last year". ciscoSystems says "here, this might crash your terminal server and void your TAC onsite agreement, but, then again, it might work flawlessly--do ya feel lucky, punk?". Lucent is still trying to iron out the bugs in its v.90 implementation, and they've been contemplating EOLing the TNT modem cards for years now--they've only held off, as they don't have a replacement part that works yet.
Believe me, there are far more factors in this equation than "$ISP doesn't want to fork out $20 and a few minutes to upload some firmware". Most of us just have our hands tied behind our backs because we just bought a buttload of equipment a year or so ago, and, with the economy being what it is, don't have the funds to replace all of it.
FWIW, I connect via dialup at home. I can't say that I'm waiting for v.92 with bated breath. v.90 is fine for me, and, from what I've seen, v.92 is only incremental--barely noticeable except in absolutely perfect conditions. Basically, if your phone line and ISP are such that you can achieve the highest-quality v.92 connection between the two points, you can probably get DSL, due to the stringent line quality standards for v.92. With that in mind, a lot of companies (including the hardware vendors) are pushing DSL a lot harder than v.92 for customers that want better upload speeds.
Read the EULA. You can purchase Windows 2000 seats through volume-licensing and "downgrade" the installation to NT 4. This applies to both the server and workstation editions.
Then, once you and your staff are trained on Windows 2000, you can upgrade, as you've already purchased the right (for n seats, anyway).
I don't know if this applies to shrinkwrap licenses, but that's how it is with volume licensing. Also, I think you can still get NT 4 media kits through the "easy fulfillment" program, if you're a volume-license customer.
I'm not carrying the Microsoft torch here or anything, but your statement is patently false. It is, however, what Microsoft wants you to believe, which is why they insinuate it, and only contradict it in the fine print.
I'd like to address the "just view the damned ads, you freeloading hippies" crowd.
Personally, the reason I started blocking banner ads (a little over a year ago) was because of one very specific ad--that stupid "punch the monkey ad".
It managed to crap more web no-nos into an ad than I ever though possible:
At the time, it was a very popular ad. I don't know what I was typing to into Altavista to make it trigger (LaTeX->latex? Monkeys? WTF?), but I seemed to get it every five pages, and Netscape dumping core every five pages was not conducive to my finding out this LaTeX technique, which I needed right then to finish a CS paper (I'd have used Fondren Library, but this was before the Rice campus library stayed open 24 hours daily).
So, as a temporary fix, I disabled Java (I didn't need it at the time), used a different search engine (Google), got what I needed, and then installed Squid+Cameron Simpson's Ad Zapper (once I'd turned-in my paper), and the problem went away. I could have Java as I needed it (Rice's CS departmnet loves Java. Turning it off in a web browser meant not being able to do certain coursework), and my browser didn't crash because of stupid monkeys.
The clear message I'd like to deliver is I don't mind non-intrusive advertising. In fact, most banner ads are very interesting, so long as they don't flash or titter about annoyingly, and don't stupidly try (and fail) to look like dialog boxes (looks really stupid under OpenWin). Occasionally, I click one. However, if it pops up in a separate window, if it spawns things in other windows, if it creates offscreen windows, if it crashed my browser, if it litters my hard drives with cookies, if it prevents me from clicking on your page, or if it dances around like a stupid monkey, I will disable it, and I will go elsewhere.
There are probably a lot of technically-minded users that feel the same way. I don't want to steal content--I don't have this need to remove all adverts from the pages I'm viewing (although, I will strip them out, if need to print the page). But, my computer is my computer, and if your website can't sit in its window and behave itself, you've just lost a viewer.
That should work great, provided you've got the RAM (more than 128M, on 6.5) and the CPU (R10k, or maybe R4400/250). Max Impact+TRAM is just an older version of MXI, without the XIO cross-connect between the CPU and gfx (which is limited to the Octane line, anyway). Basically, if SGI still published the periodic tables, you'd see MXI and Max Impact+TRAM on the same row--it's the same hardware on a different bus.
I wish I had one, but I'm getting an Octane/SI (to replace my trusty Indigo2 Killer Impact) on Monday! :) I don't do enough 3D to justify the texture memory, anyway--I just use it as a devel station because IRIX is a Kick Ass Unix(TM).
I've tried GLQuake on an Indigo2 XZ before, and it was pathetic due to the lack of hardware texturing--it handled the geometry just fine.
- OpenVMS on PA-RISC (not too scary, but... ewww), now that the Alpha is gone?
- HP-UX on Alpha, just to help Intel rape the last bit of dignity out of the platform?
- A resurgance of NT/MIPS with the new "Kayak Himalaya" workstations (dumb, but this is Compaq and HP we're talking about here)?
- PA-RISC in a new line of "Nonstop Himalaya" server (again, dumb, but is MIPS over Alpha was stupid, too)?
and, the scariest, and most-probable....- An HP-made server, with that <sarcasm>wonderful</sarcasm> HP support, running HP-UX (blech) on IA64 (cringe), using Compaq's horribly nonstandard system components and chassis, with HP's horribly nonstandard (and flaky) RAID system.
I just hope HP sells of the stuff they make that doesn't suck (calculators, printers, and medical/testing equipment) before they make something really stupid and tank. Or, maybe we'll just luck-out and simply continue to make the same crappy PCs--all under one roof. I think I need some liquid recovery now.Before we had an upstream that provided news to us, we used a "suck" server as a solution.
Basically, you run an NNTP server that acts just like a web-cache: it fetches the headers, and caches them. Then, when you download an article, it caches that, too. It also checks the upstream server for expiration and generally mimics the upstream server.
The only real downside (for the customers) is that all groups initially appear empty, except for a post that says something to the tune of "articles are downloading--please come back in a few minutes".
It's a really simple solution--your customers get news, and you don't consume a full newsfeed. If you know someone with clout at your ISP, you might want to point out that solutions like this exist (and, IIRC, are "Free").
That's a staggeringly good idea. Maybe we could even get our words and culture back. It's even case-worthy (IMHO, IANAL), since we can't admit that we're members of that culture without a negative response from Joe Average (think of the racial discrimination parallels).
The big difference that I see is that hackers are not members of any particular country, and the relavent bit of the Hague convention is designed to make IP laws enforceable under the rules of the owning party's locale.
But, at least American hackers would have a good chance--the US seems to love IP-law enforcement.
Linux Is Not UniX
I've usually had to include rm ~/core, as well, and call it from my .login file, as Netscape had a bad habit of crashing X on Linux and IRIX 5.
I think the really odd thing is that Netscape is very stable on the Macintosh. It sucked on IRIX, Solaris, Linux, Win32, Win16 (what didn't?), and Digital UNIX, but it really ran well on MacOS--better than your average Mac application.
....anyway.... I think the "What happens to Netscape" was intended to mean "Does this mean that AOL is no-longer supporting Netscape?". What about AOL/Macintosh, does it demand IE, as well? I know a lot of Mac users who won't even install Office because it's a Microsoft product--you think they'd install the program that they're convinced "killed the only decent web browser by playing dirty"?
Then again, there's really no reason they can't support Netscape 4 or Mozilla on platforms that don't yet have IE. Maybe this is just overblown like they typical Slashdot fare.
Linux Is Not UniX
Well, it's my personal experience, so any bias that shows through is as a result of me actually using both systems for years on SPARC hardware. To the best of my ability I've given both operating systems an equal chance, and I still have Linux, IRIX, and Solaris hosts on my network. I'm really only biased towards the IRIX boxen, though, which (thankfully) haven't entered into this discussion yet. :)
If your suspicions are because of my .sig, it's there just because I'm sick of interviewing job applicants that supposedly have "loads of Unix experience", yet sit down at an IRIX or Solaris console and cry "where's GNOME? Why isn't my shell bash? Where do I get the RPM for it? Does this version of IRIX have the new 2.4 kernel? Why can't I #include anything in <linux/>?". So, a sign bearing that sentence (with demonstrated emphasis) is a permanent fixture on my desk.
The vendors who tell me to "reboot into Windows and see if the problem is still there" don't help the morale, either, but that's another rant for another day....
I'm sure I could download the source to everything necessary for recovery, and build it statically (in fact, I have done this on my PC running Slackware), or download n packages to do it for me. The point is that this is part of a standard Solaris installation. /sbin/sh is static. /sbin/* are static.
I'm comparing the environments by what I get when I do a fairly reasonable install of each on the same platform, not by what I can download or hack-together--if we were to use that as a basis, all operating environments would be approximately equal.
I haven't heard of it, but (from reading the manual page) it seems to work at the fileystem level. How can that possibly affect the order in which the kernel detects devices and assigns them to node-numbers?
I wasn't saying that my partitions were drifting, my devices showed up in all sorts of odd orders, and they shift when you add or remove one. Solaris labels discs by controller number, SCSI ID, and LUN, so if you know that data, you know the name of the device, and attaching labels to anything is only a convenience--not a necessity for a consistent system.
Yes, emulation is slow, and incomplete. There are some SunOS system calls that are unimplemented, beacuse they require hardware features that Linux does not yet suppport. All of this really wouldn't matter if there were commercial apps for S/Linux.
I honestly did not know that LD_LIBRARY_PATH worked in Linux. I stand corrected. Thanks for pointing this out!
The buzzword standard is POSIX, now known as Unix95 and Unix98. That's the Unix standard.
To the best that I can tell, the free software untiites conform to the "I thought it would be cool to..." standard. Sort of like "I thought it would be cool to dump man pages and use info pages, even though man exists on every Unix machine on the planet, and I have to get info from ftp.gnu.org and it requires GNU terminfo and GNU readline and...". In some circles, GNU stands for "Get New Utilities".
If you know Unix, as a user, you know Solaris as a user. I came from an OSF/1 (ow) and IRIX background, and hit the ground running (because of the POSIX standards). I used Solaris and IRIX as a user and a developer long before I was had to manage the systems, and this adherance to standards made life easy both as a user and a sysadmin.
It's not an obvious case where free software wins anything, it's a difference in paradigms: do you add every imaginable bell and whistle, or do you stick by the rules and do what people expect. For some people, one paradigm works, for some the other works. I happen to like consistency--I think it makes life easier.
I'm not sure on the 64-meg minimum (in the license?), as I've done test installs on machines with as little as 32 megabytes (the minimum to load suninstall).
You must just be having bad luck with hard drives. If probe-scsi in OpenBoot can see it, then Solaris will see it, since Solaris just copies the OpenBoot device tree to /devices. I've got a mix of Sun, SGI, generic, and Compaq drives in one of my SPARCs, and Solaris 8 (obtained under the FBL) installs without problems.
Are you sure the problem isn't that you forgot to make a Sun disklabel on the drive? If the installer boots screaming about bad magic numbers and misking disklabels, you forgot to run format on your disc. When suninstall loads (I'm assuming you're not doing the webstart install), enlarge the console window, and type format. Tell it to label every disc--no need to partition, suninstall can do that automagically.
If you don't label the disc, the result will be that the installer ignores it, assuming that it's in use by a non-SunOS operating system.
Of course it's not documented, because if you bought disc drives from the evil marketroids at Sun, they'd be prelabeled, and you'd never have to fiddle with format.
Then again, you have to run fdisk manually in Slackware, too, so I suppose it's not horribly crufty.
Oh, poo. You're just ticked off because Sun switched from a BSD-style system to a SYSV-style system in SunOS 5 and then retroactively renamed SunOS 4 to Solaris 1, thinking that their customers would never notice the difference.
I feel your pain. SunOS 4.1.4 is probably my favorite OS on the planet. It's not Unix if you don't have to lay out your discs in cylinders, right? :)
Seriously, though, SunOS > 5.4 is really nice. It's finally as stable as the OS they had for years before. It's still not BSD, but it's better than a lot of other SYSV systems out there, and you can still run your old BSD-style apps, too, since they finally got SunOS 4 emulation working.