Slashdot Mirror

Yeah, it's been *available*, but not for actual use in any projects, which would be the point, IMO.

Wow, I have this source, and I can compile it non-stop for 25 days and read it, but I can't *use* any of it. It's like showing a beginning writer "Paradise Lost", and saying "study it all you want for the next month, but don't write it later, ha ha ha!".

(okay, maybe not Paradise Lost, I mean it still crashes on Win 3.1, doesn't know what the "Grey Enter" key is, and doesn't like some of the niftier VGA modes. But still...)
---
pb Reply or e-mail; don't vaguely moderate.

Okay, okay, I'll find a *different* protected mode app to use for my examples. DOSEmu already runs a bunch of stuff, though. I just wish it did sound better...
---
pb Reply or e-mail; don't vaguely moderate.

Now that the Bochs source is available, could this be used to make a library that virtualizes more of those pesky x86 instructions?

I'd be really happy if DOSEmu and Wine finally had full protected mode support. DOSEmu has great Linux FS support, and low memory usage, and can run Win 3.1 minus the protected mode stuff. If it had that, I don't think anything would stop it from running modern Windows, or more importantly DOOM!

---
pb Reply or e-mail; don't vaguely moderate.

Why wouldn't there be problems with massive memory leaks on Windows 2000? NT4 had that problem too. Windows works fine until either (a) you want to get real work done, or (b) it decides to crash anyhow. I'm just glad I deserted around Windows 3.1, for my personal machine.

I thought the telnet server for W2K was a pretty sweet add-on, although if I had a W2K box I guess I'd want to be running SSH instead... But to really feel at home I'd have to install BASH and the other UNIX tools. And then... well, what's the point, eh? I'd rather run Linux, have all the good stuff run natively, and run xdos over SSH if I ever need to... :)

It isn't hard for something to be a "tremendous improvement" over NT and 9x. I really doubt W2K is that, in all areas. It has higher system requirements and worse hardware compatibility. But that's the future for you, eh?

The companies roll out what they think the customers want, especially if Microsoft gives them a discount. And I don't know about Compaq or GE, but DELL really *IS* an Intel / Microsoft stooge. They try to sell computers with RAMBUS memory, for crying out loud!

There are many ways to define success. If Bill Gates cared about his customers even a tenth as much as he cares about his money, or the success of his company, I'd reward him for it. As it is, though, I fart in his general direction.

I used DOS since 3.2, and I still have a copy of DOS 2.0. I used Windows 3.1, and even though I didn't like it, I knew it well, got it to work, and found out what everything did. That all changed in Win '95, and it pissed me off. I couldn't figure out how a company could make their product less stable, more bloated, more annoying, less useful, and hide more helpful stuff from the user. Before, it was "edit WINBLAH.INI to make the problem go away". Now, it's "reboot the computer and pray".

So why do I use Linux? Because I don't trust "faith healing" as a valid system recovery method. It's *my* computer, and I'd like to know what its doing and why. I think my Operating System owes me that much. So Microsoft lost a power user. Ha ha ha ha ha. :)
---
pb Reply or e-mail; don't vaguely moderate.

Well, according to my post, your pay scale should be somewhere between $40,000 and priceless, which it is.

But since you know what the big red button is, *and* know enough not to touch it, you deserve a raise!

My numbers weren't supposed to be *too* accurate, just the sentiment behind them. See, it was supposed to be funny. I need to mark those posts with "HUMOR" again...

And I couldn't get to the Dr. Dobbs article. But their primary source had a lot of cool info, and said a lot of the things we're saying here: just because there are a lot of IT professionals out there, that doesn't mean they know how to do what *you* want them to, and they aren't all necessarily qualified.

Oh, and sorry about the NT, man. I hope you like your job. I know NT Administrators that don't do much more than reboot the server when the printer queues stop working. Scary stuff...
---
pb Reply or e-mail; don't vaguely moderate.

A Temp Worker: $25,000.
An MSCE NT Professional: $40,000.
A Professional UNIX Admin: $60,000.
An Admin who actually knows what that big red button does: Priceless.

Yep, there's a shortage. That's why I'm not worried about a job.
---
pb Reply or e-mail; don't vaguely moderate.

Because that is a slippery slope.

My social security number is just a number. So is this post. So is a computerized representation of my genetic code. According to your interpretation, no electronic media is copyrightable provided it is represented as a number (which it is, a long binary one). Or at least the *number* isn't copyrightable....

Jeez, this stuff makes my head hurt. I liked it back when I thought I was buying an actual program, and not just a license to use it. Oh, and I'd never copy it, I just hit upon the right (unpatentable) combination of bits to produce a disk that's surprisingly similar! ;)
---
pb Reply or e-mail; don't vaguely moderate.

There are *so* many holes in the DVD players, formats, etc., etc. that it won't be a problem for anyone to get around these "features" the MPAA wants.

And just how legal is this "region-coding" crap, anyhow? I remember they tried the same thing with consoles and audio and failed. (anyone ever played Golden Axe 3 in America? Anyone set the 'copyright bit' on your mp3's?)

A business isn't going to sell a product that have added features that make consumers not buy it. Well, unless they have a monopoly or something. Otherwise, a competing standard without these features will win out. The MPAA will find this out, hopefully the hard way.


---
pb Reply or e-mail; don't vaguely moderate.

I agree, but be careful with those TLAs.

I suggest:

.com for DOS sites.
.org for machine code.
.txt for sites with content / Unix sites.
.exe for Windows sites.
.hqx for Mac sites.
.cpp / .pas / .bas / .asm / etc. for source code.
.tar for archives (picture / ftp / whatever)
.gif / .jpg / .mpg for porn sites.

Under my proposal, 'slashdot.org' would be 'slashdot.txt', 'zdnet.com' would be 'zdnet.exe', and 'goatse.cx' would be 'goatse.jpg'.

This would have several advantages over the current system:

- Platform dependence. Never be told you can't get the plug-ins, just browse the right sites!

- Easy blocking of sites. Just mass-block all sites with any fluff in them! Better register all your info in the .txt domain, for the kiddies!

- Nifty three-letter name extensions. Now if only we could get rid of that darn "www." thing, and limit the other part to 8 letters. This would also cut down on domain squatting, as it limits the number of valid domains.

- Reinvent the existing system. This happened when the web was invented in 1993, and look at the economy now! It's in our best interest to completely reinvent the existing system every 8 years or so to drive interest in technology.

- Get rid of newbies. Ever since Windows stopped displaying file extensions and instead has little pictures, like the Mac, newbies don't understand them. Now that they will be unable to navigate the web, content will be back on the rise, as well as sales of "Domain Names for Dummies". I'm confident that IDG will back my proposal.

Oh, and we'll need a '.idg' for IDG books. (I wanted '.dum', but you know how those corporate sponsors are...)
---
pb Reply or e-mail; don't vaguely moderate.

D00D, that is excellent, I completely agree.

(especially since I couldn't get Wine to install FF7... blah. It isn't fair, I can play the other six under Linux! :)

Rumor has it that Starcraft is playable under Wine. If I had a copy, I'd try it out, but that's more than you're going to get from Blizzard anytime soon. Does WarCraft run under DOSEmu? (StarCon2 does, yeah! :)

Loki rulez. Except that I've been playing too much HOMM3 lately, instead of, say, sleeping. :|
Oh well, it's something to do, especially since Masters of Magic will probably never have a decent sequel...
---
pb Reply or e-mail; don't vaguely moderate.

That's how the first good checkers AIs worked: the algorithm got to be smarter than the human programmer. Of course, Go is a *much* more complicated game than Checkers. But now that Chess is basically solved, maybe people will concentrate on Go, and we'll see some new approaches. AI could definitely use some new motivation.
---
pb Reply or e-mail; don't vaguely moderate.

Point taken. I wouldn't want to see a format implemented like that. AVIs and RealPlayer files do the same thing, and it annoys the crap out of me.

The first thing to do is to make sure that all of the codecs used are implemented directly in the program / library, preferably open source so they can't be taken out later.

The interesting (and different, I think) part I was mentioning would be the ability to change which "codec" is being used at whatever point in the file. VBR encoding for mp3's works sort of like this: it detects properties about the sound (don't need as much detail / less frequencies) and adjusts its compression accordingly. Of course, that's lossy. I guess PNGs can already do this too, and they're lossless, but I'd like to see more of this for general data compression, not just images.

The advantage would be having one compressed file format that's good enough for many varieties of data, and we can go back to plain data files compressed with generic compressors. (like ps.gz instead of pdf files, or xcf.bz2 instead of .psd files, but with one great compressor. I don't think we'll ever settle on a standard data file format, even though it's probably not impossible to do...)

Hey, if you're paranoid, just store your thesis as "thesis.roff.zoo", and you should be fine! ;)
---
pb Reply or e-mail; don't vaguely moderate.

That "go" question was really cool. I wish I played it well enough to attempt to write a great computer player, as this is a classic hard problem in AI, with a $2,000,000 reward, IIRC...

Of course, his "compression" is generally expressed in terms of "rules", and even some tips from image compression might help here. (recognize similar configurations, whether they be rotated, translated, etc., and adjust your strategy accordingly)

bzip2 is a really great program, generally offering better compression than gzip at least for large bodies of text. What I'd really like to see is a meta-compression format that has some heuristic to identify the type of file, and use the appropriate (optimal) algorithm. I know most modern compression programs do something like this already, (like RAR and its multimedia compression) but it's still neat. The few bits to identify the compression methods can be well worth it...

Also, hopefully those compression patches will eventually make it into Linux; it'd be great to see something like that working at the VFS level.

If it used something like LZO, there'd be up to gzip levels of compression with practically no performance hit on even a modest system. Maybe even speed improvements would be possible, due to having to read less data from the disk...

Under those situations, I'd advocate comressing swap (and even memory!) where it would help (not recently used data), and maybe merging more of that into the filesystem too...
---
pb Reply or e-mail; don't vaguely moderate.

Beta release. For developers only.

Not even a writeup on it yet, so don't slashdot the site any earlier than you have to!

It is great to see some Perl development, though.
We're one step closer to version 6.6.6!

Anyone know when they're rewriting it all in C++?
(or are those two statements related? :)
---
pb Reply or e-mail; don't vaguely moderate.

Thanks for the info!

I haven't played the game, but the N64 controller always struck me as being somewhat elaborate. If you can't do it with a directional controller and four buttons, (six buttons max) it's probably pretty complicated... Give me a standard, NES-style controller any day.
---
pb Reply or e-mail; don't vaguely moderate.

Playing the Atari 7800 (Food Fight!) for hours at a time made my hands hurt. Solution? Stop playing every once in a while.

Playing Zelda on the Nintendo for hours at a time made my eyes hurt. But after a while, I beat it, and I didn't have to play it as much.

Maybe the controllers aren't designed for really prolonged usage, but our bodies aren't either. Some people just don't know when to quit.

Read a book, guys. Go outside. But don't sue the people who make your games just because you're lazy!
---
pb Reply or e-mail; don't vaguely moderate.

I think it lasts at least a month, I haven't messed with it lately, so I'm sure my account is long gone...

Doesn't it say on the site?
---
pb Reply or e-mail; don't vaguely moderate.

I tried it out when they started the program, it was really cool! Of course it only really matters if you're a developer or an OS nut or something, but still...

Bottom line: if you've got an app that you want to compile and run, cross platform, get an account!

Oh yeah, and to all the Trolls: they've got a Beowulf cluster, too! ;)
---
pb Reply or e-mail; don't vaguely moderate.

:) Yep, I knew all the keyboard commands by heart. Of course, this was 6.0 and later. (5.5 wasn't as cool, but I still knew the commands--the key bindings were awkward, but standard at the time, ^K^B, ^K^K, ^K^V, etc., etc.)

Not only is it unintuitive, it's darn slow! Who wants to be moving that thing around? Regardless, they supported that just fine, too, and the mouse worked well with it, if it was used. But all the keyboard commands were helpfully in the menus, and the help system was great. (much better than the standard Emacs-like Info browser, IMO...)

Yeah, most Open Source projects are far more customizable than their Windows counterparts, (because you can change it, silly!) and there's usually an (undocumented) substitute for unavailable keys. (or you can bind a key to it--I'm tempted to bind my "Windows" keys someday, but I don't need anything extra, really. :)
---
pb Reply or e-mail; don't vaguely moderate.

I'll be happy when the first thing someone says about Linux *isn't* about

* Learning vi
* Learning Emacs
* "...arcane commmand line..."
* "...text files..."

My reply to this is as follows:

* GNOME Apps
* KDE Apps
* LinuxConf
* RTFM!!!!

I think if Linux came with a "Getting Started" guide that just *touched on* the wealth of tools that are available, it'd be much easier. In Windows, it doesn't come with much. Your average Linux distro comes with a WHOLE LOT of tools, and users need to realize that.

The interface design is fine. It's just as good/bad as anything else. The last known good interface I saw was the text IDE for "Turbo Pascal 7", which is why I like RHIDE. The last "innovative" one I saw was WinAmp, (and everyone has already copied that) but that doesn't necessarily mean it's easier to use.

Widgets on Linux, Windows, whatever haven't changed much for a while. (but I'd kill to be able to resize more windows in Windows, they can't design the control panel worth a damn!)
---
pb Reply or e-mail; don't vaguely moderate.

Yeah, but when it's done, plex86 will run ".com" files, as well as whatever ".ORG" assembler you used originally.

Now I'm just waiting for the ".exe" TLD...
---
pb Reply or e-mail; don't vaguely moderate.

Heh heh heh. That must be how slashdot felt when it first got popular...

I'm always amazed at the logs I get just when I post a link on slashdot. We're a bunch of trained monkeys!

Oh well, your site goes down, but you get more corrupt on the Slashdot Purity Test. It all works out in the end...

later,

Peter


---
pb Reply or e-mail; don't vaguely moderate.

Wow, those are a lot of complicated algorithms. They weren't kidding when they said that MULTICS was huge and overly complicated...

Some of these make more sense coming from a "batch-job scheduling" perspective. Of course, we still use credit-based scheduling algorithms today (because they rule!) and I can't picture using six queues(!).

Also, about the name: If MULTICS is an OS designed by a lot of people, then Unix would be an OS designed by one person... :)


---
pb Reply or e-mail; don't vaguely moderate.

Wow, you got the darn thing to work? It apparently runs on some mythical
version of Solaris 2.5.1, and it's always a pain for me to find a machine
that will run it at all... (I'm trying it on a box running Solaris 2.6,
which is what they're all upgraded to on campus now. The script for IE5
thinks its okay...)

In case anyone was wondering what "UNIX" is according to Microsoft, here
it is:

case $OSname in

SunOS) case $OSrev in
5.[567]*)
OSdir=sunos5 ;;
5.[89]*)
echo "$OSname $OSrev is not currently supported."
echo "Please visit $IEUrl for a list of supported platforms.\n"
OSdir=sunos5 ;;
esac;;
HP-UX) case $OSrev in
*.10.[23]*|*.11.*)
OSdir=ux10 ;;
esac
esac

I got past the "display server cache" on IE4, but it isn't doing much
else at the moment...

They both It don't work over a regular X connection unless I use the
undocumented '-remote' switch, and make sure the mouse isn't on the
window to start with. One time it locked up my mouse cursor, and I
had to ssh in *again* and kill it off. I got it to show the licensing
agreement, but it will consistently "Abort" under IE5 or sleep under IE4.
Oh, and it needs 17MB of RAM to do nothing, so far.

Here are the ones I tried:

Internet Explorer 4.71.1410.4 ; Copyright (c) 1995-98 Microsoft Corp.

Internet Explorer 5.00.2013.1312 ; Copyright (c) 1995-98 Microsoft Corp.

So until IE 5.0 for Unix actually *works* on a Solaris box I can use, I'm
not too impressed with it. I hope you'll understand why--it's as if
Microsoft released IE5 and it worked on Windows '95, but not on '98,
but that's okay, because who uses Windows '98, right? :)

---
pb Reply or e-mail; don't vaguely moderate.

Yeah, I'm a rabid Linux user. I hate Microsoft, because they haven't done anything worthwhile since releasing DOS 5.0. But, that having been said...

I would happily use any Microsoft software that was ported *decently* to Linux. (you know what I mean if you've used Microsoft's "Internet Explorer for Unix". Ugh.)

Unfortunately, I have a feeling that the Wine project will beat them to it. I ran Excel '97 a while back on Wine, and that stupid paperclip came up just fine. Not much else worked, though. I'm sure it's better now. Of course, there's always VMWare, but that's not even close to native! (need a copy of Windows, too much RAM, etc., etc...)

...and if Microsoft can't play fair, let 'em burn. They've been asking for it for years. I'll happily give them another chance, I just don't think they can change their ways by now. But we'll see what the trial brings. Windows 2000 will probably make them more arrogant than ever, now that they've invented a few more features from Unix. ;)
---
pb Reply or e-mail; don't vaguely moderate.

"The Whitehouse" actually looks like a very responsible porn site, if such an animal exists.

At least, it looks like that under w3m, which is an excellent (text) browser...

If you're going to register all the TLDs you can with the same name attached, why stop there? Register all the obvious mis-spellings and other things people might try too. (because for the major sites, someone always registers these, and they put up really annoying sites!)

---
pb Reply or e-mail; don't vaguely moderate.

Actually, slashdot.[cc|com|net|org] are taken. Half of them even point to slashdot!

Fortunately, .edu, .gov, .int, .mil, .nu, and .to are still available. Get them while you still can! ;)

barrapunto.org is "Open Resources", and BarraPunto.com is the Spanish slashdot knockoff...

dotslash.com is coming soon, and slapdash.org is still my favorite...
---
pb Reply or e-mail; don't vaguely moderate.

I know. It was around before we knew what squatting was. I quote myself:

Okay, so openssh.org got taken. This happened to altavista, and countless other "big names" on the web. Some guy registers "your" name before you do, so you settle for another one.

Did I say squatting? I don't think so. Without following your link, I believe my description that someone took "their" name, i.e. the name that they wanted to use and thought was rightfully theirs because they were so attached to it, was correct. This doesn't take the intent of the original domain registrant into consideration.

And "squatting" is when you're using land that rightfully belongs to someone else. I don't know if this is that good an analogy in the first place, because domain names don't "belong" to anyone until they get registered. You can't squat on land that no one owns, and you definitely can't squat on land that you yourself own!

The only thing that's evil is when someone wastes a whole domain for something stupid, when it could go to something useful. That might be the case here, but let's wait and see first.

The OpenBSD community is known for their flamewars and bad feelings on both sides of the fence: that's how it was founded. This might be another one of those stupid pissing contests. And if someone flames me for saying so, I'll consider it further proof. :)
---
pb Reply or e-mail; don't vaguely moderate.

Yeah, and through "Network Solutions", too. Man, they're pure evil. Of course, the other "real" site is entertaining too.

Note that, to my knowledge, OpenSSH and OpenBSD both have nothing to do with "The Open Group", and that group has nothing to do with actually being open... Go figure.

[whois.corenic.net]
Registrant Todd T. Fries (template COCO-21730)
OpenBSD, the REAL open group
1523 North Pierson Apt F
W. Peoria, IL 61604 USA

Domain Name: openssh.com
Status: production

Admin Contact, Technical Contact, Zone Contact:
Todd Fries (COCO-21731) todd@fries.net
+3096739259

CORE Registrar: CORE-80

Record created: 1999-10-25 08:44:41 MET by CORE-80

Domain servers in listed order:

zeus.theos.com 199.185.137.1
cvs.openbsd.org 199.185.137.3
ns0.fries.net 209.251.96.130

Database last updated on 2000-03-07 03:55:07 MET

To optimize query speed and answer correctness see the
--help option. Depending on your whois client use
whois -h whois.corenic.net HELP
or
whois HELP@whois.corenic.net

---
pb Reply or e-mail; don't vaguely moderate.

Okay, so openssh.org got taken. This happened to altavista, and countless other "big names" on the web. Some guy registers "your" name before you do, so you settle for another one.

Openssh.org looks like a harmless list of links. Of course, the intent of Mr. Alex de Joode couldn't have been that benificent since he already has freessh.org, which is no more than a list of bookmarks, as far as I can tell.

But still, ho hum. No big news here, move along. It isn't that original a name, guys. More importantly, I certainly don't want the big companies taking away our domain names because it's their trademark/copyright/whatever. A ruling in this situation would probably set a precedent that we don't want. The only option I can think of is to try to negotiate with the parties involved, which apparently hasn't worked.
---
pb Reply or e-mail; don't vaguely moderate.

That means you're 4/5 pure, 1/5 corrupt. As a long-time slashdotter, I should be *really* corrupt, not just halfway there. :)

No, Jon Katz is a moron. Although his last rant was actually intelligible, even if he was preaching to the choir...
---
pb Reply or e-mail; don't vaguely moderate.

> Here is the result of your Slashdot Purity Test.

Okay, it was a pretty cool test.

> You answered "yes" to 105 of 200 questions, making you 47.5% slashdot pure
> (52.5% slashdot corrupt).

Woah, what do you have to do to get a good score on this thing?!?! I should have lied more...

> According to the scoring guide, your slashdot experience level is: JonKatz Wannabe

Okay. Forget this test. I can't *imagine* a worse insult to a slashdot poster. :)

(Oh, and I used "No Score +1 Bonus" and "Preview" to post this.
"Gee, I bet I'll get moderated down for this..." ;)
---
pb Reply or e-mail; don't vaguely moderate.

Why don't you use a real emulator, like MESS or Bochs that actually runs the x86 slow enough?

(MESS is made for playing old games, yeah!
(it's MAME for computer/console systems...))
---
pb Reply or e-mail; don't vaguely moderate.

Beer and the BA-2 rocket are incompatible, due to their inability to perform "heavy lifting", or operate heavy machinery with too much beer. Sorry guys, no beer in space, you'll have to brew it on the Moon after you terraform it first!

In other news, The A-Team is suing, for dilution of their "B.A." trademark. Their spokesman was saying, "I pity the fool who think he can lift more than Mr. T!"
---
pb Reply or e-mail; don't vaguely moderate.

It was, actually, until it got renamed the "Lesser General Public License", and RMS started telling people not to use it over the GPL if possible.

Oh well, I like the LGPL, it's a good balance between the philosophies. TrollTech would have saved a lot of our time, anxiety, and developer cycles if they had used the LGPL over the QPL...
---
pb Reply or e-mail; don't vaguely moderate.

I guess we can go ahead with that Quake 3 port now, eh?
---
pb Reply or e-mail; don't vaguely moderate.

In related news, Junkbuster has announced ports of their popular ad-blocking software to several smart cell-phone platforms...
---
pb Reply or e-mail; don't vaguely moderate.

Don't be so hard on Tim, guys, he's just trying to help out the community and hammer out some ground rules.

I see nothing wrong with patenting innovative ideas, although I still wouldn't put Amazon's "1-click-shopping" patent in this realm. I don't like the shopping cart metaphor, and I also don't like storing people's credit card numbers on the server, so I wouldn't consider it "innovation" in the first place. :)

Patents are a tool. Like guns and computers, they aren't evil for merely existing. They can be used and abused like anything else. I think the FSF should patent any innovations they come up with, and allow anyone writing "free" software to use those patents, and offer to license them to corporations for some reasonable price, or use them as leverage, to cross-license some proprietary features into free software. (that may be the only way The GIMP ever gets the features it needs, for instance...)

But I'd prefer it if the patents are of the innovative variety and not of the "rename an old concept now that it's new on the web" variety. (Drat, I should have patented "e-business" before anyone ever thought of it!)
---
pb Reply or e-mail; don't vaguely moderate.

I'm actually learning Scheme in Theory of Programming Languages. It's a pretty nifty course...

How do effect layers work? Is that just like an extra layer that specifies an effect?

Yes, The GIMP always allows however many undos/redos, I love that feature.

And image "slicing" annoys the hell out of me, if you mean what I think you do...

Hackers often add in features that only hackers could love. Maybe that's why people are programming with GTK...
---
pb Reply or e-mail; don't vaguely moderate.

Shouldn't that be "Anyone who says that Photoshop is more powerful than The GIMP doesn't know Scheme"? :)

But seriously, it sounds like The GIMP isn't there yet in terms of CMYK production / printing, and that's a shame. But I don't think scripting is its weakness...

You're right, though, I haven't looked at Photoshop 5.5. Photoshop 5.0, to my untrained eye, looked like Photoshop 3.0 with a lot of extra useless filters. (read: filters that I could have simulated in Photoshop 3.0 with just a little bit more work, or could have created in Photoshop 3.0 IF it had had a decent scripting language then...)

I'm also very happy with how well The GIMP and ImageMagick save files. PhotoShop, in my experience, creates bloated images that don't look as good, while both The GIMP and ImageMagick create slender, optimized PNGs and JPEGs.

Also, I like being able to save a big image as ".xcf.bz2", and have The GIMP do it automatically for me, whereas PhotoShop usually capitalizes my extensions, and *still* doesn't know which format to use unless I tell it explicitly...

But, different strokes for different folks...
---
pb Reply or e-mail; don't vaguely moderate.

Hmm. That sounds like a lot of work, especially for a home network.

I guess the initial goal would be to design the network such that the other machines get updated from a central one, (or they all mirror each other in some way) or put all the important stuff on the networked drives.

I'm sure there are software packages that could help you out here. I'd probably do something funky with locate and cron, but maybe something like AutoRPM would work better.
---
pb Reply or e-mail; don't vaguely moderate.

"art, bart, cart, dart, eeeart... Sounds good to me!"
---
pb Reply or e-mail; don't vaguely moderate.

Oh, gotcha. I wasn't thinking...

My mom is a screenprinter, and she uses Pantone colors in Photoshop and CorelDraw. I could never be as picky as she is about shades of color, so I guess that level of detail is needed in the business.

However, she also likes The GIMP 'cause it can make nifty rotating planets. I guess it just brings out the inner web designer in all of us.

Give us a nice printer, and we computer geeks can't tell the difference. :)

And isn't it CMYK?
---
pb Reply or e-mail; don't vaguely moderate.

I haven't messed with the APIs involved, especially not the Windows one. I know DirectX exists just to bypass it, though, which also doesn't sound like that sound an idea...

When X differentiates between a 'local' user (i.e. on console) and a 'remote' one better, I'm sure performance will go up. And the GGI(/KGI) people have been working on getting a fast, generic graphics interface with as little kernel code as possible, too. Hopefully that'll get folded into the framebuffer code, to the point where it supports my cheap graphic cards. :)

I also haven't used BeOS much, but from the little I know, it sounds like their advantage was redesigning from the ground up, which is something any new OS can usually afford to do. Hopefully it won't go the way of the Amiga too soon, and I'll definitely install it once it's 'free', and I have some extra space. (although I'd rather do it with a multi-processor system, to really test things out...)

What you're describing sounds kind of like a micro-kernel approach to graphics. It shouldn't be too hard to implement something like that, maybe with MkLinux. Interesting...
---
pb Reply or e-mail; don't vaguely moderate.

Yep. Once either X or the applications support pretty, anti-aliased fonts this won't be a problem. Until then, though, they'll just look ugly.

I don't have a problem with most fonts I see, but some pages look pretty bad in Netscape (I have it set to only use my fonts now) and the fonts in The GIMP look *way* better than the rest of X (they're well-rendered, and anti-aliased). The word processors in X look pretty painful, but usually I just use them to add formatting, so it's no big deal.
---
pb Reply or e-mail; don't vaguely moderate.

Hmm? What does it do currently?

I admit, I've used The GIMP to mangle a lot of images, but I don't try printing things that much.

Doesn't it just send PostScript to the printer, or whatnot? It looks like it's sending at least 24 bits per pixel... (8 each for R, G, and B)
---
pb Reply or e-mail; don't vaguely moderate.

I'll have to respectfully disagree with you on point C there. Sticking files "all over the place" is a strength, not a weakness.

RPM's exist to keep track of where all the files go. Programs are installed in the path (generally /usr/bin or /usr/local/bin) so that users don't have to make "shortcuts". (although it'd be easy to type in the name of a program and make a shortcut because the binary is in a *standard* location)

There are packaging methods that do what you describe, though. They stick the application files in a separate directory, and symlink the needed places in the file tree to there. So it has one place for the application, Unix is happy, and it has your "shortcuts" too. You can do things that way, but you certainly don't have to.

Applications install *libraries* into \WINDOWS\SYSTEM. (it's actually more complicated than that, "\WINDOWS" is really whatever the Windows root directory is set to, but...) These libraries can have multiple versions with the same name, and an *application* can overwrite a needed library that might not be binary-compatible with the new one!

This is one of the biggest flaws in Windows, which Windows 2000 will hopefully help to fix. However, the Unix method of using symbolic links to keep track of library versions, and only allowing an Administrator to install new libraries (that might conflict with the system-wide ones) is *definitely* a good thing.

If you lock users into one model and way of doing things, and only teach them that, sure they'll get used to it. And if you give Linux a default, consistent look-and-feel, people will get used to that too.

Heck, Linux is configurable enough that you could create and distribute a version of it that implements your own braindead file hierarchy, make it look Windows-ish, disable logins and only start X, etc., etc. And some people might even love it. It wouldn't be as powerful as standard Unix, but at least it wouldn't crash. :)
---
pb Reply or e-mail; don't vaguely moderate.

I agree, Windows tends to be faster due to its driver model and graphics support. It does graphics in the kernel, which is dangerous, and fast.

The rest of this sounds like application issues. First, if you have to use GNOME, try using sawmill for a window manager. Heck, I'm still waiting for GTK to mature. If you want to compare to Windows '95/'98, use something like qvwm for a window manager, it's probably closer in functionality and appearance to Windows.

Netscape under Linux (and UNIX in general) is pretty bad. To mimic the Windows experience, try using Mozilla. It'll crash more, but not as hard, and it's faster. Also, compiling stuff with "-O9 -s -fomit-frame-pointer"...etc. is always fun. :)

(or you can use Wine and try running the same applications, that might actually be more fair. That should bump the RAM requirements for Linux up to something equal for Windows, and you can compare apples to apples (almost). Linux should still be more stable, but Wine will run less stuff.)

What's your graphics card? When I buy my next system, I'm making sure all my hardware is blazing fast under Linux...
---
pb Reply or e-mail; don't vaguely moderate.

Wow. You haven't even been a user for that long, and already your default posting score is 0. I'm impressed. I guess people don't like Archie Bunker anymore... But enough about credibility.

Staroffice and Applix use MS documents well enough. My biggest complaint about StarOffice is that it's just too #*&% much like Office!

MagicPoint is also supposed to be a good PowerPoint replacement. (and PowerPoint is a *horrible* application! It can't even store images decently; I find myself using HTML and JPEGs instead...)

NeXS is a pretty cool spreadsheet, too. But there are tons of spreadsheets out there, and Excel never really impressed me that much. (Just another spreadsheet, except that it has a DOOM-style easter egg built-in. WTF?!??! That's like if Lotus 1-2-3 had a scrolly in mode 13!)

It saddens me that so many people use Office. Fortunately, 90% of the world doesn't use a computer. There's still hope.

GIMP rules. Of course, it's free, so the same people who already have PhotoShop can still use The GIMP as needed. (for its cool scripting language) The GIMP might not be as cool as the Fractal Image Painter was, but it's way cooler than PhotoShop. PhotoShop hasn't had a decent new feature since PhotoShop 3.0. (which runs great under Wine, by the way...) However, PhotoShop is also a >$500.00 application. Good thing all the Windows d00ds have already pirated it...
---
pb Reply or e-mail; don't vaguely moderate.

Linux has *tons* of great Office applications. It just doesn't have "Microsoft Office" yet. And whose fault is that, really? Also, the way wine is coming along, expect Office to work better under Linux anyhow... Of course, we still have WordPerfect, StarOffice, Applix, and even a lot of Linux-specific office suites that even Windows doesn't have... (and for real work, nothing beats TeX. :)

Preinstalled boxes don't have funky hardware issues, because the vendors pick the right hardware. I've tried to install NT on machines that could be sold preinstalled with NT, and I still couldn't configure it! Why? Because they probably got it to install *once* through black magic, and saved that image forever after for the rest of their installs. There's no shame in having a "supported hardware" list when the vendors don't do all of your work for you. And even so, Linux *still* supports more hardware than NT does...

All that having been said, it sounds like SuSE is doing a great job, and I'm very happy with SGI's commitment to Linux. Even if they're doing it out of self-interest, I don't care as long as it benefits the community as well. And that's a lesson that Microsoft should have learned 20 years ago, before they alienated their power users... If they had, and admitted their shortcomings up front, maybe the community at large would be more sympathetic towards them.
---
pb Reply or e-mail; don't vaguely moderate.