The collected works of ianj - Page 1

Somehow I missed the Solaris Express excitement here... I've been using it as my desktop OS (on an Ultra 24) since the day it came out. I've been drawn more to Sun lately because I'm not one of those people can can live completely within the abilities of my 600MHz SGI box, and Solaris to my knowledge is the only actively-developed "real UNIX" that offers modern desktop amenities. Got Express running across the board here... on the Ultra upstairs, on a Blade 1000 for SPARC builds, and on my 420R in the basement. I'm just using XVR-100s for SPARC graphics because it's readily available and does the job.

A lot of desktop-related OpenSolaris bugs were fixed around snv_148, so this release is so far proving more workable for me than its predecessors. OpenIndiana is still doing their thing, but I'm not looking too closely into them yet because they haven't released a SPARC version.

zmttoxics wrote: Did you have the special effects on? I have a feeling Osol / S11X still don't have the best video performance. I haven't run it on a physical machine yet so I can't compare...

Performance is pretty good for me, and I've got a relatively mediocre QuadroFX 370. I only use Nvidia cards and the binary Nvidia driver on my x86 *nix, though, so I can't speak to anything else. I've been sticking with that formula because Nvidia's driver is dependable and easy to configure.

I thought of Mac OS X almost right after I posted that, although I wouldn't say it's moving closer to other UNIXes - NetInfo is gone, but replaced by a new OS X-specific system, not standard /etc/passwd and so on. X11 is included, but X11 applications integrate very poorly with the Mac GUI, so it's a moot point. It is still a UNIX, of course - after all, Solaris has at least as many idiosyncrasies of its own.

I own more Macs than I do anything else and I'm definitely fond of the platform, but I've been gradually edging away from it because of, as you mentioned, the hardware. Even low-specced Mac Pros from four years ago still start well above $1,000 on the used market, but I bought a maxed-out Ultra 24 for $600 in October. I have a few grudges about the consumerization of the Mac platform, as well as a desire to spend more time in a traditional UNIX environment, but the hardware is largely at fault. Right now Solaris is where I'm finding shelter, but we'll see how things develop.

You generally need proof of residence in Japan to get cell phone service, as I understand it.

Another one of the 80mm fans in my Origin 300 went up this week. They're standard size, but use an unusual connector, and last time this happened, splicing it onto another fan didn't work. If anyone has O300 parts machines laying around, or has extra fans, I'll buy as many as you can offer.

Some other related items I'd be interested in:

Origin 300 mounting rails (unlikely, I know, but I'm going to keep asking)

SGI PCI Gigabit Ethernet Card

1GB RAM cards for IP35 machines

Any Fuel R16000 CPU module

V12 graphics for Fuel

jan-jaap wrote: I replaced the fans in my O300 so it can be done: viewtopic.php?f=3&t=16718040
I still have the original fans, but I'm in Europe.

Did those Scythe fans have the proper connector? I saw your post earlier, but the "BUY NOW" button on the site you linked to doesn't do anything, and I can't find that model for sale anywhere else.

jan-jaap wrote: Gigabit ethernet: but a cheap 3c996bT or equivalent Compaq/HP card and hack it, instructions found here: viewtopic.php?f=3&t=12702

I'll take a look into playing with that, but it looks a bit beyond what I have the time/patience for. Won't rule it out until I try it, though.

recondas wrote: There's also a photo of the wiring from an SGI 3-pin fan connector spliced into the wires of a non-SGI 3-pin fan <that shows the wire color mismatch> at the bottom of the Fuel (IP35) Hardware Aggregator: viewtopic.php?f=3&t=9046

Cool, I'll try wiring one of the off-brand 80mms I have like in the picture. I've only lost one fan, so the system is still running with the warning light on and I should be able to use l1cmd to check the RPMs. That being said, I'd still be interested in buying some bona fide O300 fans, even if it means shipping from Europe... I've ordered heavier merchandise from that part of the world before.

pierocks wrote: Ask and you shall receive...I just checked out my O300 and it's got a completely standard ATX 3-pin fan pinout. The connector isn't the regular IDC, but the pins are in the same order (1->BLK/GND, 2->RED/12V, 3->YEL/TACH). I even verified this with my o-scope to make sure a tach signal was coming down the yellow wire. By the way, here's the digikey page for the fan in the O300, it should help you find a direct replacement.

So the wire colors match up on the O300?

Also, I did stumble upon that digikey page when searching for replacement fans last time this happened, but never found anything beyond it.

Japanese society is very welcoming of foreign visitors, but you will meet a lot of (generally subtle, unspoken) resistance if you try doing anything along the lines of actually living there. This little idiosyncrasy is a helpful thing to keep in mind while conducting business in the country.

zafunk wrote: I'd guess that less than 5% make it to the highest levels. Likely because they had some ludicrous motive for learning Japanese, like understanding anime/video games/dramas/music lyrics, or finding a Japanese girlfriend. I would say that you should ask yourself: "Why do I want to learn this language?" It's going to require one heck of a lot of effort to learn and gain any real fluency.

This. I've been studying Japanese at varying levels of intensity since I was eleven years old, and most of the classes I've seen had a fair contingent of people sitting in the back, listening to J-Pop and tracing pictures of their favorite anime characters. For all their enthusiasm for Japanese pop culture, they struggled with the material and constantly complained about how hard it was. Learning Japanese is a key to understanding Japan, not just its pop culture; make sure this is what you want.

That being said, I'm going to share some of my thoughts on learning Japanese that seem to go against the common wisdom of Japanese classes I've taken -

Don't try to learn how to speak first and put off writing until later. I'm going to go out on a limb here and say that, if you are trying to learn Japanese, you should not ever use Romaji. Kana are pretty simple and should only take a couple days to get familiar with. If it takes you longer, no problem, but learn them before you try learning how to say things. The longer you put off writing, the more you'll get used to thinking of the language without it and the harder it will be. I've seen a lot of people have particular issues with Japanese writing, and I think much of the reason for this is because they keep trying to avoid it. If you embrace it from the beginning, it will be easier for you. This also means learning Kanji. When you write, use as many as you can and pencil in the kana for the ones you don't know. Learn about the structure of Kanji and the radical system instead of trying to memorize them as monolithic units (the way many Japanese classes try to do it). Even if you are more interested in learning to speak Japanese, take some time to develop reading and writing as well. If you can read Japanese, you also get far more opportunities to immerse yourself in the language.

This won't hit you right away, but when it comes time to learn about verbs, don't learn the polite forms first. Every Japanese class I've been in has done it this way and it confuses the hell out of people. You will be confronted with a lot of learning material that tries to throw -masu forms at you first, which makes all the other forms of verbs look completely random when you go to learn them later. Japanese verbs conjugate based a highly consistent pattern of endings that is generally not explained in its own right in classes. I had a chance to teach a class that was having trouble with verbs, and this was all it took to clear them up. This will probably look like going too far ahead and making a huge deal out of a small part of the language, but if you can't conjugate verbs, you're obviously going to have big trouble. Unless you're facing a situation where you need to carry on a basic conversation in public immediately, take the time to learn things in a logical way instead of starting with the selected bits that a textbook writer thought you might need first in the wild. You're going for the real deal, not preparing for a business trip.

My overall point is to seek out the patterns underlying everything and learn using them as a guide. Too many Japanese classes (and language classes in general) try to start by pinpointing a few specific things and hiding the context because it is "too advanced," with the end result being that everything is hard to understand because little to no context is provided for anything. Pick up a few different books, search the web, etc. instead of relying on one author's idea of how you should learn things. And whatever you do, steer away from materials that try to teach you in Romaji.

There are a lot of great web-based resources for learning Japanese. I've just started digging around this site, but Tofugu is very well-designed and so far, I think the author has a good grasp of language-learning. Here is his "100 Best" list of Japanese learning resources that could help get you started.

Here is a good video adapter to get if you want to sidestep all the monitor compatibility headaches:

http://www.deepspacecables.com/models/octane.html

I'm fortunate enough to have SGI monitors and cables, so I don't have any myself, but I set up a friend with an Octane and one of these is working well for him.

I run an O300 and appreciate this information. I'd say it's worth waiting for the system to show up before buying any, though - I have a feeling many O300s were run as nodes in multi-unit setups, never had hard drives put in them, and hit the recyclers with their factory drive sled/air baffle inserts untouched (like mine did).

Not so fast... we will need a WebM-capable application first. This announcement opens the door for IRIX because unlike Flash, such an application can be ported (or written), but someone has to walk through that door, which is going to be a major undertaking. Any takers? A better web browser is what the SGI community needs more than anything else.

skywriter wrote: I always found the web to be a pointless distraction to enjoying an SGI with IRIX.

If I want to reference something online or grab the URL for a source package while I'm working on an SGI, it's much easier to do so on the same machine than turn to another machine and copy it over. Firefox 2 is generally okay for this, but it's not that great either.

Generally I'd agree that introducing new formats is not necessary, but in this case we're talking about replacing Flash, which only runs on platforms Adobe feels like releasing it for, with something that anyone can port to their platform of choice. That's definitely worthwhile.

mila wrote:

skywriter wrote: I always found the web to be a pointless distraction to enjoying an SGI with IRIX.

Must really agree on that, I did not get my Indigo to just surf on that I can do a a Solaris box BUT it is however nice to be able to surf from a IRIX box.

But if we take the latest browser or any software and manage to port it I doubt that any IRIX box can handle the load, the old MIPS boxes are to slow, I know how my Sparc boxes are running firefox and the gnome crap and they are 1.6GHz machines with pretty good graphics.

Receiving H264 or similar on a IRIX will that really work or do we need a coprocessor also?

Eh, the SGI I use most of the time runs at 800MHz. Seeing how Firefox 2 performs on it, I don't anticipate a more capable browser being a problem, nor do I think faster SGIs should be held back by concerns about the abilities of their '90s predecessors. Yes, an Indigo is not going to be able to do much with a current web browser, but that doesn't mean such a thing has no place on IRIX as a whole. While I don't go to my SGIs with web browsing specifically in mind, I do things on them that involve a web browser, whether that be looking up documentation or grabbing the URL for some source code, and something other than an old version of Firefox would make that process more pleasant.

As for SPARC, I have a 900MHz Sun box in the other room that can handle Firefox just fine (that includes Facebook, Youtube, and other such fluffy entertainment sites). The "if it can't do it as fast as a brand-new x86 PC, it's not worth it" argument gets old.

IBTL

OH MY GOD, HE DOESN'T LOOK LIKE ME!

OH MY GOD, HIS NAME DOESN'T SOUND LIKE MINE!

HE CAN'T POSSIBLY BE A REAL AMERICAN LIKE ME!

Grow up, guy.

This is another useful guide I've used. Between it and the wiki's "Installation to Fuel" article I've generally been good; just be careful to replace everything machine-specific in the examples with what's relevant to you.

As for installing from a local hard disk, I'm not sure it will let you do that. The firmware-based installation tool expects to load distributions from either a CD-ROM or network source, but maybe you could choose CD-ROM and point it to the hard disk? I haven't done a non-network install in ages, so I can't remember.

You will learn to love IRIX network installation. I Guarantee It.

Cares enough to continue providing IRIX patches. Good enough for me.

If you can't find something in the standard repository, SFE is the first place to look. Blastwave et. al. are severely out of date and were never intended for anything newer than Solaris 10. OpenIndiana is hosting an IPS repository of prebuilt SFE packages, which you should add:



It contains boost packages, but not opencv. Try building it yourself from the SFE spec. Also, keep in mind that pkg list will only print packages installed on your system. You'll need to use pkg list -a to print the entire contents of the repo.

Also be aware that the location and organization of oi-sfe might change. OpenIndiana hasn't finalized how they want to handle packages yet.

josehill wrote: Nice to see built-in support for whole disk encryption with an "instant wipe" capability. The new system restore options are also welcome, including a Windows-style restore partition. Speaking of Windows, Lion includes some Windows-like features, like a new "Resume" function (seems like Hibernate on steroids) and the ability to resize windows from any window edge.

I don't like the addition of full-screen applications and window resizing from any edge. These are major identifying features of Windows, and I don't think they belong in the Mac OS. Full-screen apps aren't as big of a deal because you don't have to use them, but the prospect of accidentally resizing windows when trying to move them is not a good one. Apple will probably do any-edge resizing better than it's been done before, but I still think it's a bad idea.

josehill wrote: Most significant of all, however, is the aggressive move to the cloud. At the moment, Apple has no plans to offer Lion via DVD -- if you want it, it's a $30, 4 GB upgrade, installable from the App Store. Will people really want that, or will people insist on DVDs? Will this mark the real beginning of moving all of a user's computing experience to the cloud, abandoning the device-centric computing model of the last few generations?

I'm not interested in putting my data in the "cloud" (I have my own servers; why keep my data on somebody else's?) but what Apple is doing there is going to be a huge gain for many users. The installer is more iffy. I don't care whether it comes on a DVD or not, but I like to apply new major OS X releases as fresh installs, not upgrades. It's important to me that the OS be available as some sort of full-install image and not just an update pulled from the Internet by a running OS X installation. I assume some image will circulate at least so repair techs can reinstall it on machines, but whether it will be available to the public is another story. We'll see.

Apple is moving farther away from traditional desktop standards, which shouldn't be a surprise to anyone. I don't like all of the directions they're taking things in, but I'm mostly going to watch from the sidelines and give them the benefit of the doubt, at least at first. I've been shifting away from OS X and towards more "normal" UNIX on the desktop, but I travel with a MacBook Pro and may get an iPhone next year, so I intend to keep a foot in the Apple world.

My laptop is 32-bit and therefore won't run 10.7, so, again, watching from the sidelines.

ChristTrekker wrote: I sometimes would appreciate the ability to drag any edge to resize, but I still believe in the original Apple HIG of "least surprise". Resizing a window is more "destructive" than moving it. Maybe option-drag or something to resize would be fine, though.

This is exactly what I was thinking (resizing should be considered too destructive for the entire window border to trigger it). Apple has a history of using modifier keys for less commonly-used window operations, so see your suggestion being a better fit than changing the default behavior. While we're on the subject, Alt-drag to move windows from anywhere is the UNIX window manager/desktop feature I miss the most when using OS X. That would also be nice to have considering how the upper parts of most Mac OS windows are increasingly taken up by live UI elements.

ChristTrekker wrote: I don't know if a converted-AAC MP3 would be as good as an originally-encoded MP3, but it may be possible.

Transcoding is possible, but nobody should ever do it. Ever.

Just had to say that.

There is no such thing as a "reference release" of Mac OS X. You all are crazy.

If you have conservative tastes in computing environments and want to be able to use a given UI with a reasonable expectation that it won't change with version bumps, proprietary Unix and BSD are waiting with open arms. People being disappointed about something together on the internet happens all the time, but when it gets to be about topics like this, there tends to be an implication lurking in the shadows that Windows and Mac OS X (maybe Ubuntu if the young fogey in question is feeling particularly adventurous) are the only options, which is not only false, but would be hilarious on a forum of this nature.

commodorejohn wrote: Yeah, and that's great, if you like Unix .

If you don't like Unix, what are you doing here?

(couldn't resist)

As for OS X 10.6 vs. newer versions... I've been through the 10.6 -> 10.7 -> 10.8 upgrade cycle, and I have no idea what people who think something went bad after 10.6 are talking about. Unless you're like the OP and think the smaller window controls look "cheap" (huh?), or you really, really care about having colored sidebar icons, that is. But really, if you're going to obsess over the details of your UI to that level and insist that they never change, you should be using a different platform. As far as I can tell, the actual user experience is the same across all three versions.

6.0.8 was the best version of the System Software. Damn those developers fumbling around with the latest trends like colors and beveled buttons. User-selectable desktop pictures? Nonsense! A computer should just do the job it is expected to do. It is a tool and nobody is allowed to have fun with it or think it's pretty.

I think the only reason anyone needs Windows or Mac OS specifically anymore is that are gamers or they work with Microsoft Office documents for a living, and generally if you do, you do it on a machine provided by your employer, not your personal box. Same goes for any other commercial software one might use on the job.

commodorejohn wrote: You didn't answer my question. How is it "terrible or dangerous," as opposed to just "not what Microsoft wants?"

It would take developer time to maintain the special, just-for-commodorejohn legacy UI, continue porting it to future versions of Windows, and ensure that future applications are compatible with it, which means Microsoft would either need to pull developers from other projects just to work on it, or hire people for the task. How much are you willing to pay them for the privilege?

guardian452 wrote: found the facebook/twitter integration with lion (or ML?) annoying, since I use neither and found it a bit insulting at first. I think that was the tipping point for me to become a greybeard geezer even though I don't consider myself that old or having been around all that long. https://web.archive.org/web/20110207204 ... m/science/

You have a fair point in that Apple is narrowing the markets it wants to focus on, but I find this particular bit absurd. I don't care for Facebook either, and I have never noticed the Facebook integration in the latest Apple operating systems because I have not gone looking for it. If there is some unavoidable change to the UI that gets in your way, by all means, talk about that, but feeling insulted by an optional feature that you will never see without deliberately deciding to use it is crossing a line into ideological craziness. That's like saying Mac OS 9 is insulting because the Setup Assistant asks you once, on first boot, whether or not you'd like the enable Simple Finder (for those who don't remember, the default is "no").

commodorejohn wrote: Well, first off, it wouldn't be just for me, as the fact that XP still has something like 25% market share nearly thirteen years after its inception and seven years after it was no longer the flagship and now after it's been officially and very publically EOL-ed attests. But let's say that's true, and there literally is nobody else but myself who wants this: all I'm doing is kvetching on a forum, for chrissakes. I'm not staging a commando raid on Redmond and taking Satya Nadella hostage until I get what I want. Ineffectual? Sure, if we discount the catharsis factor. But "terrible or dangerous" is just typical Cory5412-brand ludicrous hyperbole.

I believe he said using actually Windows XP on the internet was terrible and dangerous, not the complaining itself.

As for your 25% figure, that's worldwide, isn't it? Most of that is likely accounted for by pirated copies of XP in use in developing countries and by businesses that dropped the ball on the upgrade process, not by people who are refusing to upgrade to Windows 7 because they don't like the interface. Either way, I suspect Microsoft has far more detailed information about whether they would make or lose money on a "legacy interface" option than you do, and they're not doing it.

Yeah, sure, you've kvetching on a forum. One thing about forums is that they are full of other people who might not agree with you.

commodorejohn wrote: Oh, I don't mean to imply that every single one of those millions of users have only the UI holding them back from rushing out to buy the latest version of Windows - but come on, man. Even if it's only one out of five of those, that's how many people?

If we assume that one out of five Windows XP users are not upgrading because of the UI (and I think that's being very generous), that's still only 5% of Windows users worldwide. You would have a very hard time convincing Microsoft that it is worth the expense to maintain an entirely separate UI indefinitely for 5% of its userbase, and again, I think the actual figure is much lower.