Slashback: Encumbrance, Silence, Internalization

Slashback with two different updates on the donation by Sun of elliptic-curve cryptographic techniques to the OpenSSL project, the state of Microsoftization of the U.S. Department of the Interior, and the strange outcome of Batt vs. the Cage Trust. Read on below for the details.

Different folks, different contributions Dr. Sheueling Chang-Shantz writes:

"Hello, I am the lead researcher/developer of the ECC project at Sun Microsystems Laboratories. I appreciate very much the news you posted on Slashdot regarding 'OpenSSL Gets Cryptography Gift From Sun.'

However, your wordings "Sun Microsystems has donated ... developed by Whitfield Diffie ..." seems to be causing some confusion on Slashdot forum. It gave the wrong interpretation that Whit has invented ECC. Sun is definitely making no attempt to claim that Whitfield Diffie has invented the Elliptic Curve Cryptosystem. Technically, neither has Whitfield Diffie developed the ECC technology that Sun has donated to the OpenSSL project recently.

I would appreciate it if you could correct the news before too late.

For clarification, Elliptic curve cryptography was independently invented by Neal Koblitz, Professor of Mathematics at the University of Washington and Victor Miller who was then at IBM.

Whitfield Diffie is Sun's chief security officer who co-invented Diffie-Helman public-key cryptography."

We now go north of the border ... And further on the topic of that donation by Sun, friscolr writes "In a recent post on misc@, OpenBSD project leader Theo de Raadt states...

OpenSSL is becoming a non-free software project, because the code from Sun contains licenses which invoke patent litigation; the licence on the new code basically builds a contract that says "if you use this code, you cannot sue Sun".

He goes on to say, 'once again, i think it is time to fork OpenSSL.' Thank you, Theo, for always making sure we will have 100% free software at our disposal and for standing by your stated goals."

[Headline redacted] Dotnaught writes "The question of whether British composer Mike Batt's "A Minute's Silence" on the "Classical Graffiti" CD (by The Planets) violated the copyright of John Cage's silent composition " 4'33" " has been resolved in an out-of-court settlement. Batt reportedly paid the John Cage Trust an "adequate sum" (whatever that is). On his site, Batt writes, 'We have now settled the matter of my artless plagiarism of John Cage's silence, by his publishers caving in and us winning! Why didn't I think of that before! We could have saved a lot of time and buggering about, although I must say, the struggle was one of the most amusing disputes I've ever , er, disputed.' Batt may yet have the last laugh. According to the New Yorker, Batt has been busy copyrighting chunks of silence of various lengths other than the four minutes, thirty-three seconds of silence owned by Cage."

Hey, does this guy really work for the government? In response to broadly worded news that the U.S. Department of the Interior was switching to an all-Microsoft computing infrastructure, security architect (and oftc.net honcho) D. Clyde Williamson fired off a well-phrased mail to Hord Tipton, Acting Chief Information Officer for the Department of the Interior. asking for clarification, and urging that the DOI consider advantages of not tying themselves completely to proprietary systems. Tipton's response (posted with his permission) is informative:

"Thanks for your views on the DOI's attempts to standardize operating systems. Whereas it is true we are moving towards enterprise approaches to desktops and operating systems, there will be as you suggest a heterogenous mix at the server level. We have not decided at this point to be 100% Microsoft although that discussion has been entertained. There are certain risks and efficiencies that must be considered regardless of the path taken.

Our major concern is interoperability and our current situation is all over the map. Thus standardization is an important step forward for us.

Thanks again for your views.

Hord Tipton
Department of the Interior"

Why relying on a single vendor for such an important aspect of the modern workplace is still considered an "enterprise approach" I'm not sure, but it is certainly true at many companies.

Comments

At the client level

[Dancin_Santa]

It makes a lot of sense to have everyone using the same operating system at the user level. Standardize the OS, disallow unapproved app, device, driver installation, and use an OS that doesn't require extensive training.

Heh, you'd think they'd go with Mac.

Re:At the client level

[mmol_6453]

(And just to clarify your point)

That's "at the user level."

They're still leaving the door way open for running different types of servers.

I'm rather impressed at the prompt response of a major player at the DOI. What with all the requests for press he's probably getting, he appears to have a great deal of store set in relatively private "public relations."

Could someone give good, logical reasons? I'm seriously all ears.

Re:At the client level

[Anonvmous Coward]

"Heh, you'd think they'd go with Mac."

I know you meant this sarcastically, but you inadvertently touched on an interesting point: The more interest you have with your computer, the more efficient you'll become with it.

I'm really good with Windows. Always have been. But when I got my first job as an animator, they put me on an Alpha station running NT 3. (yes 3... or was it 3.52 or something like that? All I remember is that the interface resembled Windows 3.0, and I was used to 95.) My boss suggested I find some plugins for Lightwave and get them installed. But I was afraid to mess with this thing! Not only was the interface really different, but it also had an entirely different processor. If it had been NT4 (Umm.. not quite sure if NT4 was ready to go then...) I would have been pretty comfortable in playing with it. Why? Because I used Windows 95 at home and the interface was similar. I had a pretty good idea of what I could do with it and not feel like I'm going to break it.

My point? Well, it's safe to assume most of the people there have a Wintel PC in their house. If the computers they use at work are Wintel as well, they'll be more comfy with it. No matter how good an OS is, it is difficult to support somebody who's unfamiliarity with their system makes them scared to mess with it.

Re:At the client level

[ealar dlanvuli]

To be quite honest, everything from MS should run on a citrix box, everything else should run on apropriate X servs. Macs need local programs (to the best of my knowledge). That way you have no ties to the desktop OS at all, and you can jump to Xenizen (the imaginary OS I just made up) tomorow so long as it supports java (citrix client) and X.

There is no reason anyone except programmers should ever run a program locally. And all PC's should be a simple ghost that allows them to run citrix apps.

Re:At the client level

[evilviper]

No matter how good an OS is, it is difficult to support somebody who's unfamiliarity with their system makes them scared to mess with it.

Back in the good old MS-DOS days, my High School had a Menu system in place on all their machines. (BTW, I heard their MS-DOS days only ended about 2 years ago :-) ) That menu system was incredibly simple. Use the arrow keys, or the mouse to highlight an option, then hit enter, or click the mouse, to run that option. Even if you've never seen a computer before, you could get on one of those systems, and instantly master the art of formatting disks, writing a document (plain text, with a line of instructions across the bottom), and too many things to name.

Now, the question is, what made those systems so much easier to use than today's systems. Some will say complexity, but I don't think so. I believe that the lack of descriptive text is what kills the computer. I'm not sure who to blame, Xerox, Apple, Microsoft, but the model that was set, was coppied without question, and here we are today.

So, when someone sits down at a new computer, the colorful icons, and arbitrary names mean nothing to them. It just happens that the more savvy among us may recognize the Netscape icon, and understand how we use it. To a newbie, it's no more descriptive than a red triangle icon labeled 'Fred' would be to you.

So, my solution, as I've said it before and will say it again, is to include A LOT MORE TEXT in interfaces. Just imagine how easy it would be to use a new computer, with a desktop full of icons that explained themselves. At a glance, you would understand what to do with the lighthouse icon: "Double-Click To Look at Web Sites". (tooltips/baloons aren't the same, and don't work)

How about GNOME & KDE? Their 'Start' buttons are just arbitrary pictures, that mean no more than anything else on the panel does (as shown by Sun Microsystems' GNOME UI study). Now if they had text below them that said "Main Menu", everyone would know right where to go.

This isn't limited to window managers either. You just have to guess what a B might mean. Now, if word processors had "Bold Text" on the toolbar, anyone could figure it out in an instant. Icons are still important, as they make good shortcuts. If you know that bold has a blue square next to it, then finding the bold button is much quicker after repition.

Of course, text does not automatically solve all problems. Take a look at menus. A "File" menu has misc stuff under it. What is needed, is logical organization, then some logical wording along with it. Obviously, a "Text" menu would be a very good start for a word processor.

One other thing to end this rant... Another absolute DON'T is clearly illustrated by Windows. You should have only ONE WAY to do something. So, if you want to delete a file, going to File-Delete should be the end of it. (Keyboard shortcuts are exempt). If you tell someone how to do something, a different way then they have done it, they might think they are doing something entirely different: "No, I didn't 'delete' it, I 'removed' it."

So, my point in all this, is simply that interface design is not an enigma, nor a price you have to pay to use a computer. It just seems that people design UIs as they've seen them designed. So, if someone was to make a GUI from scratch, keeping this, and other simple design considerations in mind, I'm convinced that a neanderthal could sit down, and almost instantly feel completely comfortable working on, configuring, and maintaining their system.

Re:At the client level

[evilviper]

all this text would take a lot of place on the screen.

Umm, yeah, because we all know how much space Netscape took up with a text description for their buttons (Back, Forward, Reload, Home, etc).

Without choice there is no room for personal preference.

Give me a break. Just because you can't choose between two different menus through which you can delete a file, you feel that it would somehow cripple you? Maybe by only giving you one keyboard shortcut to perform an action, we are again destroying your preference...

A lot of people fail to realze that a computer is a tool, not a thing that magically does what you want it to.

A computer is a chunk of metal, silicone, et al. it doesn't do much of anything. It's the software that does the job, and yes, you can make the software much more simple for the people using it. That's what the whole category of UI design is about.

People that can't be bothered to learn the basics of operating a computer (I'm talking desktop GUI here, not command line and advanced configuring) don't need a computer enough to learn to use the tool.

That's some interesting hypocritcism there. What if I was to say that those not willing to learn how to use a computer from the command-line, should be using the computer? How about that if someone isn't willing to learn how to rebuild their engine, they shouldn't be driving in the first place?

Those statements have no more merit than what you've said.

However, that's besides the point. The idea is not to put thousands of animated characters in front of you, it's simply to do a better job describing the function of the items on a computer, to make that computer easier to learn and use.

Your arguement could be taken to extremes both ways. You could say that anyone who isn't willing to dedicate their life to learning everything their is to know about a computer, shouldn't use a computer in the first place.

I don't think anyone in the real world would go along with you on this one. There is no limit to the ammount of money spent on training classes to teach people how to use a particular piece of software. If someone was to make an equivalent piece of software, which was so inherently descriptive that far less training was needed, they could charge an amazing ammount of money, and still, every company would want to buy it.

Re:At the client level

[evilviper]

Left out something...

What if we went to the extremes in your direction:

"Text takes up too much space, so we should remove it all together. Now all you have are icons to identify each program and function.

At the same time, we could stick an icon for every single function on the toolbar of Word. Why deal with those pesky menus that slow you down?

And you know, it takes longer to distinguish icons with pictures, so we'll just limit it to sold colors, and basic shapes. So, now the blue circle is what you will click on to launch Netscape. Of couse you will need to read through our 500PG manual to discover what each color+shape icon does. But hey, if you aren't willing to spend 2 years memorizing each and every function, you shouldn't be using a computer. Just like, if you are not willing to build it from scratch, you should not have a refridgerator.

Oh, yes, to open Netscape Setup, you single-click. To launch the Netscape editor, you doublick. To start Netscape with a single window, you triple click. Hey, it's more to learn and memorize, but it'll save time, since you don't have to go through all those pesky menus."

Re:At the client level

[evilviper]

I prefer a middle ground.

I understand. I was merely explaining the extreme ends of your example. Showing ways it could be made more difficult for you or me, was an attempt to give you a little better perspective on the issue.

Let's keep using a wordprosessor as an example.

Good idea. Looking at AbiWord, I could put descriptive text, using up about 50% more screen space. Now that's assuming you want to keep it layed out the way it is. Of course, even if you wish to argue that speed is a more important issue, in that text we could have the keyboard shortcuts, which would actually lead to an increase in speed. Besides, just like Netscape, we can always have an option to disable the descriptive text, but being enabled by default (so a new user will benefit from it) is the important part.

But there will have to be sacrifices. Power = complexity.

What I'm saying is that I don't believe that is true. If the icons on your desktop have more or less text, does that make it any less or more powerful?

I could argue that VI/Emacs could be made as simple as pico. For instance, with the GUI version of VI, you can easilly access all the functions, without having to learn the key combinations. Now, VI isn't perfect, but it is still a good example of the overall idea.

Re:At the client level

[mmol_6453]

Even if we realize that nobody else cares, what difference does it make?

We still disagree vehemently with what they do, and that's the first step at changing it.

I'm speechless

[Greyfox]

The fact that you can copyright silence renders me speechless.

But only for 2 minutes.

Re:I'm speechless

[EvanED]

Are you sure those two minutes aren't copyrighted?

Re:I'm speechless

[aardvarkjoe]

No; I'm pretty sure that it was actually a derivitive of my composition, "Twelve Minutes Without Connecting the Microphone." I'll have my lawyers contact him soon enough to arrange for a reasonable fee.

Re:I'm speechless

[stuffman64]

Sorry, but all of that work is derivitive work of my "3810 Minutes and 50 Seconds of Absolutely Nothing at All."

Ok, its really just my spindle of 50 blank 74 minute CD-Rs.... But still, you all owe me a "reasonable fee!"

Re:I'm speechless

[aardvarkjoe]

So, what did you have in mind ... couple million? That might leave you a dime or two when the vult^H^H^H^H lawyers take their cut.

Re:I'm speechless

[phil reed]

You ought to know that the John Cage piece 4'33'', which is the length of the piece of silence, is actually made up of 3 movements of 30 seconds, 2 minutes 23 seconds, and 1 minute 40 seconds. The score consists of an appropriate number of pages of (empty) music, and the performer is to signal the end of each movement.

A history and discussion of the piece can be found here.

Re:I'm speechless

[EvanED]

I'm sorry, but I hae trumpted you all: I have "6048 Sad, Silent, Lonely Hours" under my name. (This is 36 weeks * 7 days * 24 hours BTW)

Re:I'm speechless

[gosand]

I wonder if I could copyright the digits of pi in binary?

Sure you can.

Of course, copyrighting an infinite sequence of non-repeating numbers may be difficult to define, but go for it.

Re:I'm speechless

[Sir Tristam]

I'm sorry, but I hae trumpted you all: I have "6048 Sad, Silent, Lonely Hours" under my name. (This is 36 weeks * 7 days * 24 hours BTW)

You seem to be infringing on my seminal work, The Song of the Universe: A Sixteen Billion Year Retrospective. I'm still looking for a publisher of the 105,193,000,000,000 80-minute CD set, so if you know anybody who's interested, let me know. It really starts off with a bang, and the next few hundred million years are a seething malestrom of pure energy, eventually resolving to a more sedate work. Your "Sad, Silent, Lonley Hours" seem to match a section of my work that started 137,518,824 years 15 weeks two days five hours ago (give or take a little).

Chris Beckenbach

(Is the horse dead yet?)

Everyone must post

[PD]

I have copyrighted the act of NOT posting on Slashdot. If you don't post, you're in violation. If you don't post twice, you're OK. I haven't copyrighted that. As far as I know, that one's under the GNU copyleft.

Re:Everyone must post

[Tablizer]

I copyrighted a server not running.

Never mind, Microsoft has too much prior art.

I like Free Software and all.

[/dev/trash]

He goes on to say, 'once again, i think it is time to fork OpenSSL.' Thank you, Theo, for always making sure we will have 100% free software at our disposal and for standing by your stated goals."

So if Theo or any other 'major' player hadn't said Sun was making OpenSSL non-free and to fork it, we'd still use the Sun OpenSSL?

Re:I like Free Software and all.

[evilviper]

Well, you have a wonderfully obsecure way with words, so I'm guessing your meaning here.

I would say 'yes'. I can come up with plenty of examples where an I.P. problem with a single portion of code resulted in a whole being on shaky legal ground.

4.xBSD-lite & 386BSD comes to mind right away.

In other news

[spun]

I have copyright on various lengths of passing gas. You may not issue a fart of 3, 4, or 7 seconds without violating my copyright.

Re:In other news

[cei]

Uh huh. The infamous "Seven Seconds of Silent but Deadly..."

Re:In other news

[Tablizer]

I have copyright on various lengths of passing gas.

Well, your lawyers are welcomed to stick their face near my ass to inspect the duration.

Waiter, another Burrito Grande, please.

Re:Yeah, right

[EvanED]

Right on! They should be 100% *nix! Why don't they see the light?

Facetiousness aside, they're considering it because they should investigate all the alternatives. 100% MS is a viable option, albeit a poor and risky choice for most applications, but a choice nonetheless. One should investigate all the alternatives before coming to a conclusion.

Re:I hereby claim the copyright on...

[Anonvmous Coward]

"I hereby claim the copyright on......all posts not submitted regarding this article."

Plagarist!

Every word this person said has been written before! Here..

Re:Yeah, right

[susano_otter]

Why not? There are advantages to a homogenous environment. Many of these advantages are the same no matter which vendor provides 100% of your systems.

But hey! Let's consider the "alternative": 60 webservers all serving the same site, some running IIS, some running Apache, some running Iplanet. Now, go and maintain all of that.

I work in a very heterogenous datacenter, but all machines of the same type, in the same environment, run the same code on the same platform. The reasons for homogeneity on some level should be readily apparent.

This man is not who he claims to be

[delta407]

FYI, according to the OpenBSD site it's "Theo de Raadt", not "Theo DeRaadt".

Don't believe me? Check this user's posting history, Theo's personal homepage, interviews, or mailing list posts.

Re:This man is not who he claims to be

[delta407]

Sure, it's a common mistake, but on Slashdot you sign up for your own account. (Do you spell your own name wrong?) It's obvious from the Google results that "Theo de Raadt" is the preferred spelling (31,000 vs 1,800) and there is nothing related to his account that would validate his identity.

I stand by my conclusion.

Re:This man is not who he claims to be

[billn]

Having met the real thing, based on your observation (the likening of Theo to a huge dick), it could be the same man. Don't get me wrong, he's a brilliant kind of guy, and he does great things with oBSD, but if he's got more than a shred of social skill, I'd be fairly amazed.

no need to fork OpenSSL

[plcurechax]

In the cryptography mailing list, it appears that Theo may not need to declare jihad on licenses he doesn't like.

According to Ulf Möller there will be a patch made before the next release to isolate the ECC code in case of patent concerns. The ECC code can be included or excluded based on a configure flag like the present RC5 and IDEA algorithms which are still patented in various parts of the world.

Apparently the patent claim is an additional optional provision that companies can use the Sun code under a truce against lawsuits if they agree to not sue about ECC patent infrigement either.

Re:no need to fork OpenSSL

[stebilad]

According to Ulf Möller there will be a patch made before the next release to isolate the ECC code in case of patent concerns. The ECC code can be included or excluded based on a configure flag like the present RC5 and IDEA algorithms which are still patented in various parts of the world.

Compile-time flags already exist to turn on and off ECC code in OpenSSL - they are OPENSSL_NO_EC, OPENSSL_NO_ECDH, and OPENSSL_NO_ECDSA. Additionally, there's a compile-time flag to turn on or off the code that is allegedly encumbered by Sun patents and a compile-time flag to turn off code that might be encumbered by another company's patents.

Furthermore, this is not new to OpenSSL nor to the crypto world in general. Lots of algorithms included in OpenSSL are covered by patents, RC5 and IDEA being prime examples. The OpenSSL license and most other open-source licenses only give you rights to copy and distribute the code, not necessarily to use it. Just as it was illegal to use RSA cryptography in the United States before Sept. 2000 without licensing it from RSA Security, so too is it illegal to use RC5 without licensing it. The OpenSSL license does not and cannot grant you those rights.

The Sun provision is there to grant users additional rights. As the previous poster indicates, it allows you to use any algorithm that Sun has a patent on in the context of OpenSSL and be free from threat of patent infringement lawsuit provided you don't sue Sun over a related issue.

Is it reasonable for Sun to ask you to not to sue them for code they gave away for free in return for not suing you? That's a business decision you make when you decide to use OpenSSL code.

Is it reasonable for Sun to say you can use the encumbered code in the context of OpenSSL but not in other contexts (like a hardware accelerator)? Under US law, they've got the right to do that. Whether you agree with patents or not is a different argument.

Sympathy...

[Anonvmous Coward]

"We have not decided at this point to be 100% Microsoft although that discussion has been entertained. There are certain risks and efficiencies that must be considered regardless of the path taken."

Like or hate their decision, anybody who's ever tried to print from a Linux box to a printer hosted on a Windows machine can sympathize. Technical superiority is fine and all, but ease of use has a larger impact on overall efficiency.

Re:Sympathy...

[jbolden]

I think you are a little out of date. Recent versions of Samba + Cups make this trivial. What distribution / version are you using?

Re:Sympathy...

[miffo.swe]

Its not hard at all to print to a dozen of windows boxes if you want to. Just put up a SMB printer proxy that relays the linux boxes to the windows boxes. That said i cant remember having seen anything in windows that makes printing from windows to a unix server easier either.

In a network controlled by proffessionals such "problems" are nobrainers in comparison to security and stability. In linux ease of use is only limited by the administrators imagination.

Re:Sympathy...

[ealar dlanvuli]

Most companies have at most 5 printer models, and all of them accept Postscript.

That point is really not valid at a corperation, though linux does have some drawbacks to other OS's, Windows only has the advantage of being "vaugly familar" (tell that to the gal who couldn't minimize windows I talked to earlier today..) with alot of driver support.

In a corperate setting, especially ones that run *alot* of CITRIX, other solutions are (or at least should be) always being considered.

To be quite honest, no one does anything except reghost a machine if it's more complex than "my printer is pointing to the wrong JetDirect box", so therefore the cost of support is how often does 1: the printer get redirected 2: how often do you need to reghost. There are no other costs for support client side (users are capable of handling prety much any windowing system with standard titlebars, so thats also fairly moot).

The real benifit of other OS's is you can cleanly stop users from saving anything locally, this is a major advantage when it comes time for the next reghost and someone dosen't loose 5 hours worth of work.

Re:Sympathy...

[redcliffe]

It only took me two clicks on the KDE printer wizard and I had mine working. Don't know what the heck you must have been doing......

Re:Sympathy...

[phliar]

... anybody who's ever tried to print from a Linux box to a printer hosted on a Windows machine can sympathize.

FUD alert! I don't understand this at all. How long ago was this situation you're talking about? Under Mandrake 8.2 (and possibly other distributions) if you use "new lpr" as the print system, printtool or the "Control Center" will set up printing to Windows -- just type in a few parameters into a dialog.

Re:Sympathy...

[brunes69]

Use KDE and KDE apps. All printing problems ceased for mt wonce I switched to CUPS/KDE. Everything is a breeze to install and configure, including printing to windows machines over a LAN.

Re:Sympathy...

[dasunt]

apt-get install cupsys cupsys-bsd
[ Wait for download on 56k ]

lynx localhost:631
[ Setup my printer - a simple Deskjet 540 ]
[ Print test page - CUPS is now working ]

vi /etc/samba/smb.conf
[ Set printing = cups ]
[ Set printcap name = /etc/printcap.sys ]
[ Set load printers = yes ]
[ Set print command = lpr -P %P %p %s ]
[ Set a few other minor things ]
/etc/init.d/samba restart

Download postscript driver
[ Wait for 56k connection - Could have used one of the apple postscript drivers that came with 98, btw ]
[ Install network printer under windows ]

That was hard. Think I need a beer. :)

Re:Sympathy...

[dasunt]

Under 2000, there is the option to install 'Unix Printing'. A download gives 98 this feature as well. It will print directly via CUPS on port 631 of the machine in question.

Re:Sympathy...

[jbolden]

If you are using "apt-get" you are using debian and aren't a desktop user. Try it on Mandrake

1) ghostscript is already configured as a filter
2) mandrake control center interfaces with both cups and samba so that the windows printer is now configured (which I assume you are doing the setup above only makes sense if you want to share off the printer which wasn't the actual question)

Re:Sympathy...

[jbolden]

Look even if Windows can see the printer that doesn't mean its going to print anything. That doesn't mean much. The fact is for most printers a distribution like Mandrake is no harder then configuring a windows printer:

a) where is the printer
b) what kind of printer is it

Re:Sympathy...

[AJWM]

This has got to be the most irritating pending issue for linux to make it on the desktop. Even if CUPS can see the printer, that doesn't mean it's going to actually print anything.

And this differs from Windows how?

I just upgraded my wife's machine. She runs NT (some Windows-only apps). It promptly decided it could no longer see the laser printer on the network. At least, the print software decided that. I can ping the printer. I can log into its admin software. But but the Windows software refuses to acknowledge it's existence. I finally just gave up on that crap and connected it via a parallel cable (the printer's in her office). The Linux boxes and Macs in the house have no problem with it over the net.

Re:Sympathy...

[jbolden]

I phrased that very badly by dropping a word;
it should read "If you are using 'apt-get' you are using debian and aren't an average desktop user". In context the conversation was about using cups for printing. I was saying that on distributions like Mandrake setting up cups for simple print is no more difficult then setting windows printers.....

I don't know where Debian stands in terms of tight Cups integration; though I'd assume it wouldn't be a default for them.

Now on your major point about all major Linux distributions being roughly the same regarding their desktop support because of similar software I disagree 100%.

1) Debain focuses heavily on making sure the system works virtually the same on many different platforms (sort of like NetBSD) this is totally useless, and worse harmful for a desktop user who rarely needs this kind of platform conformity.

2) Mandrake focuses heavily on creating very easy administrative tools to allow naive users to self administer the system. For dedicated servers under high quality administrators this is useless and possibly harmful.

3) Redhat desktop, Mandrake... work hard to get the desktop to function out of the box perfectly so it is quite large. Debain by contrast makes the desktop highly configurable especially in terms of size and extensions.

I think if you were to take even more extreme examples like:
Knopptic or Lindows vs.
Rock Linux, Beehive Linux, Securepoint linux

These all have the same software but you are cutting against the grain.

Anyway the main point was that IMHO a Debian user should be able to install cups and the easier distrubitions do the work for the user. I think you would agree with that.

Re:Sympathy...

[HiThere]

This is correct for a locally attached printer. For a printer on a network ... there are all sorts of considerations. (I finally resorted to using TCP/IP to link to my network printer, but not all of our printers have that kind of connection available.)

Also: On the windows machines, available printers are auto-detected, together with the name assigned to them by the sysadmin. On linux, you need to track down the printer, determine it's parameters (which won't include the name that the sysadmin has given it), and then enter them by hand. Possibly this is because we are a Novell Netware shop, but Netware has been around for a long time, so this shouldn't be the sticking point. (OTOH, perhaps when we upgrade to v6 of Netware things will improve. Perhaps.)

Re:Sympathy...

[HiThere]

It's pretty easy for most locally attached printers. If the printer is out on the LAN though... It's not necessarily easy at all. (At least not with the version of Netware that we are using.)

Re:Sympathy...

[HiThere]

You may be describing where you work quite well, but it doesn't match where I work. We have perhaps 40 printers, and on Windows, at least, there are a multitude of different drivers (10? more?) I suppose that they can all accept postscript, but that will usually result in the special features (which is why we got the special printer) not working. So the approach that you are proposing yields at best a kind of minimal common printer, which is good enough for me, but not for most of the people who work here (appearances are considered *quite* important!).

Also, many of these printers just aren't visible to Linux systems at all. The Netware queues aren't useable, so one needs to use direct TCP/IP addressing, and not all of the printers have that capability/enabled. (JetDirect assumes one particular kind of connection, which is true for some of the printers, but not for others.) (I think we use a bit of CITRIX, but certainly not much!).

Perhaps the sysadmin could change this. And only shared files go on the servers. (We keep running out of space.)

Part of the reason for this is historic. (For years the network would go down at times for reasons that no one ever discovered. We eventually decided that it must have been a hardware problem, because at some point it just stopped happening [hard to pin down exactly] shortly after a round of upgrades. And we still don't know what was [occasionally] bad.) Anyway, so the users had to be at least minimally able to operate with the network down. And they still like that. But windows printer access depends on the Novell network. And on Linux, it appears necessary to use only those printers with TCP/IP active.

Re:Sympathy...

[Anonvmous Coward]

"It only took me two clicks on the KDE printer wizard and I had mine working. Don't know what the heck you must have been doing......"

Ah yes, the "My experience accounts for everybody's experiences" style of debate. Heh.

Re:Sympathy...

[jbolden]

I can't speak to netware but for pure windows printers all you need is printer name.

DOI Duckspeak

[Black Copter Control]

The reply from the DOI pretty much says almost as much as the 4 minutes of silence that was being used as a PR toy.

A "heterogenous mix at the server level." could simply mean a mix of NT2000, NT4 and XP. Although one could hope that it really means other manufacturer's systems as well, it doesn't have to.

For the rest of it, it sounds like they still intend to force the desktop to pure MS.

Re:Yeah, right

[delta407]

Yes, but he was arguing for interoperability, not consistency. I'm all for consistency, whether it be M$ or *nix. Of course there are advantages to a homogenous environment, but if you go homogenous Microsoft, you're tied to Microsoft, period.

If you run *nix, you can use NFS or a variety of new network filesystems. If you run Windows, you get SMB. But wait -- there's Samba for *nix that lets Windows speak its own little proprietary protocol and interoperate with *nix servers. *nix can speak Windows, but not vice versa.

If you run everything on Apache, you can host your sites on Linux, Windows, Solaris, FreeBSD, what-have-you on a variety of different architectures. If you host on IIS, you're stuck with Windows and the very limited number of platforms it supports. Apache runs on Windows, IIS does not run on *nix.

If you write your website in PHP, you can use it on a variety of Unicies on a variety of different platforms. If you write for ASP dot NET, you're stuck to Windows 2000+ on x86 (and whatever else Redmond feels like supporting). You can run PHP on Windows, but not vice versa.

So... tell me, which is the more flexible solution? Which delivers more interoperability? The open, freely extendable system or the closed and proprietary one? That's why I pointed this out as a no-brainer; it is.

(And yes, I know about Microsoft's UNIX tools, but it's a moot point.)

Say what?

[Kwil]

A seven second fart?

The heck with violating copyright, that sounds like it violates physiology.

Re:Say what?

[alienmole]

Nah, you just need the right diet... Your dinner tonight.

An adequate sum

[wfmcwalter]

Batt reportedly paid the John Cage Trust an "adequate sum" (whatever that is)
Apparently Batt gave the Cage Trust a suitcase full of no money.

silence

[Satai]

Ok, this is important to me. Yeah, it sounds stupid that the suit was over silence - but what it really was about was that he credited Cage as an author and did not pay the estate. THAT caused the problem. Even Sonic Youth did a track of silence and didn't get sued - because they didn't have the cavalier audacity to credit someone else without checking the ramifications.

So how about we stop making fun of the situation? Cage's estate isn't at fault here. That guy shouldn't pull such stupid shit.

Re:Silence

[jpt.d]

I think they missed out - God probably patented it, copyrighted it and made it a trademark before any of them were born. Now that the patent and copyright has run out (5 billion years out...) they are public domain!!!

Re:He seems to be a reasonable facsimile, then

[delta407]

• His "e-mail" is just a mailto link in his .sig, not one on his username -- if it was linked from his username, I would believe you. You can't fake that, you have to have that e-mail address to get the account. Anyone can put a link in their .sig
• His "journal" is one entry. 'Nuff said.
• His point may stand, but everything seems to point that he is not who he says he is. (Besides which, why would he post at zero if he's such an insightful guy?)

Well then...

[Eric_Cartman_South_P]

...I'm not posting, just to prove a point! Oh wait... shit...

Re:Yeah, right

[afidel]

There are several nfs drivers for windows, there are asp drivers for several web servers other than IIS. Basically there are at least partial implementations of almost all important technologies going both ways. The real goal should be interoperability through published standards.

Next Step:

[Bistronaut]

Next Step: Outlaw all sarcastic humor.

Watch out, Onion, you're on the hit list. Cuz' I'm pretty sure that Bush didn't actually threaten to invade the West Nile in response to the West Virus.

Hello? Sarcasm? Where did that go?

I wonder that too

[mindstrm]

Is this an actual proven fact.. or is it just the same common misconception people make... that because seomthing goes on infinitely, never repeating, that means every possible combination must exist somewhere. (not true)

You can't copyright that

[hayden]

It's not a piece of work and so can't be copyrighted. It's a method and so needs to be patented.

Email me for a licence on "Method and apperatus for disseminating a plurality of absence of content via online bitching servers".

Thanks.

Re:You can't copyright that

[foobar104]

No, you misunderstand. He's not copyrighting the method of not posting, he's copyrighting the act itself.

Re:You can't copyright that

[EvanED]

You can't copyright an action, a method of acting, or anything similar. You can only copyright a tangable work, such as a piece of music, a movie, a book, whatever.

On the other hand, you *can* patent the method of not posting, though there's more prior art than you can shake a stick at, so it may not get through the USPTO.

Re:You can't copyright that

[DEBEDb]

Sure you can. Actual performances can be copyrighted (not recordings of them, or
plays, or musical scores, but those one-time
performances themselves). Which is why
recording an orchestra performance in a
concert hall is a violation. You are not
copying a recording, and the original music
(Mozart, say) is long since in the public
domain.

Phew.

[Wolfier]

You should all be thankful to the /. lameness filter that filters out empty posts otherwise you'll all be in violation!

Re:copyright of pi

[Bishop]

PI is an ifinite sequence of non-repeating numbers. Every finite sequence of numbers can be found within the digits of PI. It may take a while to find your finite sequence, but it is there. This is similar to how all finite numbers are contained within infinity.

As the digits of PI have considerable prior art, I would suggest that the digits of another irrational number be copyright. Among other prior art from PiDigits we see:

The National Energy Research Scientific Computing Center maintains a web page in which binary-encoded words (with a = 1, ..., z = 26) can be looked up in the first 4 billion digits of pi.

The page goes on to list some other interesting sequences of numbers and their positions.

Batt's settlement

[rsidd]

Batt reportedly paid the John Cage Trust an "adequate sum" (whatever that is).

He paid them a six figure sum.

Re:Batt's settlement

[dmiller]

It should have been $000000

Re:New Business Model?

[Tablizer]

1. Copyright 3, 5, 7, and 14 question marks.
2. ???
3. ?????
4. ???????
5. ??????????????
6. Profit!!!

Something to look forward to

[Lucas Membrane]

See:

http://www.angio.net/pi/piquery

Not only does every possible finite sequence of bits occur in the bits of pi, it occurs an infinite number of times. It's all there: directions to Jimmy Hoffa's grave, the human genome, the lost works of Shakespeare, MPEG's of Gallmer and Bates doing the unspeakable, a bug-free release of Windows 2010, JPEG's of those court and military records Bush won't release, MS-DOS 1.0 with Gary Kildall's Easter Egg still in it, everything! An infinite number of times! Find it. Post it here.

Re:Something to look forward to

[Hard_Code]

Well, the MPAA and RIAA better get their drones up to Congress to ban pi.

Re:Something to look forward to

[hysterion]

Ahem. From D. Bailey and R. Crandall, On the random character of fundamental constant expansions, Experimental Mathematics 10 (2001), p.276:

"Even the weaker assertion that every finite digit string appears in the expansion has not been established, to our knowledge"

Re:Something to look forward to

[mindstrm]

That's interesting.. but nowhere there does it show any kind of proof.

"Finding any string somewhere" is not an inherent property of nonrepeating infinite sequences of numbers......

Re:The situation at the Interior Department

[alienmole]

When I am talking all Microsoft environment, I mean Windows 95, because that is what he is running on a P3 machine down there, and it came with windows 2000 on it.

Are you trolling? You can't be serious, and more importantly, you can't possibly be correct. I can't imagine a department-wide policy, in 2002, being formulated that involves standardizing on Windows 9x/Me over Win2K or XP. Windows 9x/Me is based on MS-DOS, for crying out loud! The admin problems are legendary! Oh, the humanity! Oh, the tax dollars!!

enterprise approach

[novarese]

Why relying on a single vendor for such an important aspect of the modern workplace is still considered an "enterprise approach" I'm not sure, but it is certainly true at many companies.

Ah, grasshopper, you've just labeled yourself a novice. The reason you're not sure why that's considered an enterprise approach is that you have no experience with enterprise-class operations. You can get a vendor to agree to all kinds of massive price reductions on hardware and, more-importantly, the margin-laden services contracts, by agreeing to standardize your entire operation around their products.

Better to NOT Give Due Credit?!

[Myriad]

...because they didn't have the cavalier audacity to credit someone else without checking the ramifications.

Excuse me? Let me get your reason straight here...

It's okay for me to pinch something, so long as I don't give due credit? /Me thinks that's pretty damned twisted.

My experience has been people are generally much happier with you if you DO give them credit for something they came up with. Usually it gets nasty when you try to pass their idea off as your own.

Don't get my wrong, the idea that you can copyright silence is ridiculous. Your statement strikes me as even more so.

Re:I hereby claim the copyright on...

[jdbear]

Why are you so blatently harrassing elderly, weight challenged children of single mothers who were able to rise above their unfortunate births to accumulate a little wealth to comfort them in thier old age? The poor things, having to live with the shame of being born illegitimate, growing up fat (undoubtably do to overeating in reaction to thier low self esteem.) You should be ashamed.

Re:Yeah, right

[cscx]

Wow, tell me that was written without bias. Windows can speak NFS. There are hundreds of utilities for just this. Just like you have to install smbfs support in Unix/Linux, you have to install it on Windows.

If you write your site in ASP (note you cite ASP.NET, to further your agenda with a brand new product), you can run it on Windows, Unix/Linux, and NetWare. Ever heard of Chilisoft! ASP, or Novell NScript?

Your idiotic points are akin to saying "Windows binaries don't run on a 4-way Sun box! Huzzah!"

When you can play the field fairly, and present valid points, feel free to do so. Otherwise STFU.

BTW, "more flexible" does NOT mean "more capable" or "more productive."

why would you want users to mess with work PCs

[lordpixel]

Rightly or wrongly, I think many helpdesks would be happier with machines that users leave well alone!

Much time, money and effort is often put into locking machines down and making sure the users are anything but comfy.

Not that that's a reason to recommend one OS over another, but your point can be taken both ways.

Re:I hereby claim the copyright on...

[dhanav]

Batt reportedly paid the John Cage Trust an "adequate sum"....

Well this proves that "Silence is golden"... atleast when copyrighted ;)

Re:copyright of pi

[Bishop]

Yes you are correct! Mod -1 against me for not thinking it though.

Incidently I get lost on a tangent following the links from Wolfram. Once again I am reminded that math is wacked. But in a good way.

Alternative to OpenSSL

[zdzichu]

Despite fact OpenSSL is so widely used, there exist a project to make GPLed replacement for it - GNU Transport Layer Security Library.

It is useful for all those people, for whom BSD license is not enough free. I think that TLS (the new name for SSL, BTW) library is mandatory for GNU/Operating System. And because of GNU it has to be GPLed - now it means reimplemented from scratch.
I also fear, that it will be binary incompatible with OpenSSL - if so, it wouldn't gain popularity. It should be drop-in replacement.

But we will see - right now you can test it or go and help developing this crypto library.

NetBSD take on the OpenSSL / Sun issue

[LizardKing]

I pointed out OpenBSD's concerns with OpenSSL on the NetBSD security list, and later summarised the points being made by Theo and others. The subsequent debate highlighted the fact that this is not a copyright issue, but a patent covenant one, and that Theo et. al. had misunderstood the purpose of Suns comments.

The hope is that the Sun code will be moved into a dedicated directory, as has been done with the problematic idea code. Then the code can be omitted when building binary packages for release. The source can be shipped with the offending code, and the end user can recompile OpenSSL to add it back in if the patent covenant is not an issue for them.

See the NetBSD mail archives at http://mail-index.netbsd.org/tech-security/2002/09 / for details.

Chris

Re:silence

[MeerCat]

If I wrote a book and released it as "Written by MeerCat and George Bush", or J. Frantzen, or J.D. Salinger, or anyone else famous, do you think they would have reason to complain ?

How about if (knowing the way I write) it was a poor rip-off of one of their writings, but was also volume 3 of an 8-volume set that was complete and utter shite, and slanderous and pornographic. Don't you think I'd be trading off their name and unjustly trying to claim authority for my work ?

Now, there's nothing to stop me putting a dedication to a famous author, or claiming it's an homage to them, but if I claim that they WROTE the book with me, I guess I'd get my arse sued off endlessly.

Batt was being an idiot, and just enjoyed mis-reporting the case because it gets him publicity.

T

Re:The situation at the Interior Department

[codepunk]

Actually your dad is the idiot, I take exception to a government facility spending my tax dollars upgrading to something that is no more functional. Don't give me some bullshit about 2/k and xp being more stable since it is just no so.

God's copyright

[McFly777]

Umm... Unfortuately copyright now is valid for 75 years after the death of the author. Therefore, if God copyrighted silence, the copyright would still be valid.

Re:Silence

[DEBEDb]

In the beginning there was the Word, though.
No silence :). It's unclear whether the Word
was copyrighted either.

Re:The situation at the Interior Department

[alienmole]

I'm not aware of any MSDOS-based applications that can't be run on the NT/2K/XP platform, with the possible exception of some games and things that do funky direct hardware access, which is hardly likely to be a requirement at the Department of the Interior.

There's no way a new policy which involves standardizing on Win95 can be rationally defended for a large organization today.

Re:I hereby claim the copyright on...

[Anonvmous Coward]

"Someone has been watching "3rd Rock" recently."

Yep!

Ironically, my funny post accusing somebody of plagarism was plagarized. Heh. I was hoping more people'd catch that, though.

Fair use

[Fjord]

Would less than 10 seconds of silence be fair use?