Other fields have dark sides, too: I refugee'd out of academia twelve years ago -- they offered me tenure, which gave me the opportunity to assess what kind of political bs^H^H^Hgames I was putting up with, and I decided I didn't like it. So I found another job before April was out, then on June 15, I gave them two weeks' notice.
The whole set-up in academia is like the dark side of the LA Times article, only worse. The three-year-and-move junior faculty contracts are designed for exactly that purpose: by the time you've learned the ropes at a place, you have to do all you can just to have a good chance of getting the next job. And it isn't necessarily better for senior faculty, either.
I just finished email with Prof. Froomkin and he informs me I referenced an old version of the criticism, targeting a previous draft of the WIPO proposal. The final proposal is now much improved; two-thirds of the issues he identified have been dealt with in some degree (still problems with the last three). His preliminary critique of the final WIPO proposal can be found at http://www.law.miami.edu/~amf/firstlook.htm -- "a fuller analysis is in progress"
Have a look at this analysis by A. Michael Froomkin, Professor of Law at the University of Miami, and a Member, WIPO Panel of Experts, Internet Domain Name Process: http://www.law.miami.edu/~amf/quickgui de.htm
In short, Froomkin says the plan is seriously flawed, and constitutes a radical subversion of existing legal checks and balances:
Bias. The plan is biased in favor of trademark holders [as opposed to others using the web as a means of speech and press];
Enabling censorship. The WIPO plan fails to protect fundamental free-speech interests including parody, and criticism of corporations;
Zero Privacy. The WIPO plan provides zero privacy protections for the name, address and phone number of individual registrants;
Intimidation. The WIPO plan creates an expensive loser-pays arbitration process with uncertain rules [Plaintiff gets to choose rules anywhere in the world!! -- not just in defendant's country] that will intimidate persons who have registered into surrendering valid registrations;
Tilts the playing field. The WIPO plan would always allow challengers to domain names registrations to appeal to a court, but would often deny this privilege to the original registrant;
Smorgasbord approach to law. Instead of directing arbitrators to apply applicable law, WIPO proposes using additional, different, rules it selected-rules that will often disadvantage registrants.
Froomkin gives a link to his detailed (50-page) analysis. I think this proposal needs to be sunk!
RPM 3.0 in RedHat 6: is Bug 236 fixed yet?
on
RedHat 6.0 is Out
·
· Score: 1
It turns out that the RPM in the RH5.2 does not evaluate whether it has enough scratch disk-space to complete an upgrade before it blindly goes ahead. If it does not have enough scratch space in that partition, it can corrupt it badly. I reported this, and they designated the bug as #236 in their database. With their new bug-tracking software, I don't have a password to check the current status of this bug. Is it fixed in RPM3.0 yet ?? (There's no way I'm going to upgrade RH until it is fixed. I don't want to go through that again!
It's probably because he's using some M$ product for his writing: they persistently put in metadata claiming they're ISO Latin-1 when they aren't -- in particular, their "opening double-quote" and "closing double-quote" characters violate the standard. IIRC, there was an "Ask Slashdot" recently about this...
And it turns out to be an interesting property of Slashdot's comment form that when you preview, it takes an HTML quote-symbol and displays it literally in the "Preview Comment" subwindow but turns it into a double-quote punctuation mark in the source (Post Comment) subwindow. So "Preview" repeated twice fixes the problem...
anyone know when I will actually be able to have X working on a tnt2?
It won't be fully optimized for the TNT2 yet, but the TNT support in XFree86 3.3.3-1 or later (available as .gz's from Xfree86.org or as 's from RedHat) should do an at-least-decent job.
I love my 2400x1600@32bpp virtual displays on my current TNT; the TNT2 promises to be even better!
anyone know when I will actually be able to have X working on a tnt2?
It won't be fully optimized for the TNT2 yet, but the TNT support in XFree86 3.3.3-1 or later (available as .gz's from Xfree86.org or as RPM's from RedHat) should do an at-least-decent job.
I love my 2400x1600@32bpp virtual displays on my current TNT; the TNT2 promises to be even better!
The TNT2 chipset is actually a TNT chipset made a lot smaller (0.25 micron)
Not just smaller and faster because of the feature-size reduction: the rendering pipelines are re-implemented as well. With the TNT drivers (optimized for TNT timing), TNT2 runs up to 17% faster than TNT at the same clock speed, according to Tom's Hardware Guide, and with drivers optimized for the TNT2 timing it should be better yet (this is like optimizing for 486 vs. optimizing for P-II). Additionally, TNT2 offers support for larger (32MB) frame-buffers as well as faster 300MHZ RAMDAC -- and 2048x1536 displays@85hz
Bit by bit, we're getting closer to the true desktop metaphor (I want my desktop to be 8K x 4K @ 200dpi and 90Hz:-)
...I wonder if we will find any GPL'd code in there buried somewhere deep:) Probably not, GPL code works well, unlike some other bits of code...
I recall someone writing something a month or so ago about running strings on some of the Microsoft DLLs, and turning up some suspicious results. Unfortunately, I don't have the reference right now, and I don't have the time to go ahead and do a proper job of it right now; it would take hours to do it right, and this is the middle of proposal season (;-/ (so why am I posting on Slashdot:-))
Can someone be persuaded to try it and post the results?
When you release something as public domain--or in a general way under BSD and X licenses--you do not restrict the freedom of your users. Furthermore, by making YOUR code public domain, you also assure its forever-after availability to whoever wants to use it. The thing you DO NOT do is assure that derived works are similarly free. But the type of freedom I want is the ability to grant free use of *MY* code to others... but that does not mean that I want to *infect* those others.
I wish it were that simple. And although the case law does suggest that copyrighted derivative works of public-domain works must specify what is the copyrightable content as distinct from the public-domain content, the practice is that they don't -- and the publishers have deep enough pockets to engage in litigation-terrorism very effectively. this has long been a recurring topic on music newsgroups. A few examples from 18th-century music (which of all things should surely be public domain by now!):
Novello's edition of Handel's
Messiah: "Copyright 1959 by Novello and Company, Limited. No part of this publication may be copied or reproduced in any form or by any means without the prior permission of Novello and Company Limited. Permission to perform this work in public must be obtained..."
Oxford University Press's edition of Haydn's
The Creation: "Copyright Oxford University Press 1991 All rights reserved. Apart from fair dealing for the purposes of research or private study, or criticism or review, as permitted under the Copyright Act of 1988, this publication may only be reproduced, stored, or transmitted in anyy form or by any means with the prior permission in writing of the Oxford University Press. Permission to perform the work should normally be obtained... Permission to make a recording must be obtained..."
Walton's edition of Vivaldi's
Gloria: "Copyright 1961 by Walton Music Corp...All rights reserved."
To be fair, I should point out that a few publishers, such as Kalmus and Schirmer, do not make such grandiose claims, at least in the editions I checked.
The upshot is that I trust the Copyright Act and lawyers! a lot less than I did in my more naive days. fwiw.
JosefK writes, "Let's use a fictional illustration..."
Let's use a real one: the fact that the Gateway 2000 computer company has used this kind of trademark-litigation terrorism with respect to uses of the word "gateway" or Holstein cows is why I do not even consider them when making computer purchases. (Their trademarks at the time all included the number; they did/do not claim "gateway" by itself as an official trade mark.)
You likely will need massive memory and IO bandwidth to complement an insanely fast CPU...I can fantasize about an Alpha in an Octane or an Origin, that would be a complete screamer...
There was an article on The Register yesterday about upcoming Compaq Alpha building blocks -- modular system components that would make a real Origin-killer (bandwidth out the gazoo!! Have a look at this
While I'm not a huge fan of StarOffice, it's about as full featured as MicroSoft Office. Hardly "rudimentary".
Really...
I've got StarOffice 5 on both linux and solaris, and sorry, it just ain't so!
Any product that only allows one instance of one window into an app is brain-dead. SO doesn't even let you have 2 windows into the same doc, much less multiple windows into multiple docs. I need those multiple instances visible at the same time -- otherwise I might as well be relegated back to a DOS-style app. For all I hate M$, at least Office got that much right!
Gateway has made a systematic practice of trademark-litigation terrorism. Among other things, they seem to think they hold the ® on Holsteins;-(
You may recall their hassling the "tucows" freeware site because of the Holsteins; they have also done quite a bit of other improper trademark litigation, IMNHO. I will never purchase systems from firms which engage in this sort of practice.
They do make big flat-panels, if you're willing to pay for them. Sony makes a 30-inch 150dpi flat-panel that's mostly used for very high-res/space-constrained/price-insensitive work (it runs about $15K). You know, things like AWACS planes and medical radiography...
I wish I could afford one! Maybe they'll be affordable one day - it's progress that they are available at all.
If you buy a software medical package, and instead of going to the doctor (v expensive), you fill out some details with signs and symptoms, and it mis-diagnoses you and you get seriously ill, how happy are you going to be?
And when you go to a doctor, he asks for the symptoms and mis-diagnoses you and you get seriously ill, how happy are you? Don't say it doesn't happen; the doctors very nearly killed my father that way.
And statistically-significant sampling shows that most doctors don't even know the right way to bandage-up a sprained ankle!
...given similar L2 cache sizes and reasonably modern hardware, it's not really that much better than an UltraSPARC or high end PowerPC based AS/400 system for serious number crunching.
That is very application-dependent. Don't try running the MM5 meteorology model on a Sun, unless you want a hindcast instead of a forecast! Our benchmarks indicate that while our air pollution models run fine on Crays, DEC^H^H^HCompaq Alphas, SGIs, or Suns, the meteorology model runs reasonably only on the first 3. (Actually, we find that a 2-year-old 21164/400 machine matches the rest without even going to the latest 600+ MHz 164's or to the 264's)
They are a DOE spin-off working on archival technologies. The idea is to use particle beams to do the writing instead of lasers: you can focus the beam much more tightly, hence make much smaller dots. They have two technologies -- digital holding 165GB/disk, with 20MB/s storage rate, and analog, holding 90,000 pages scanned at 300dpi. Both use _very_ durable silicon-wafer substrates.
At that density, a 6-platter changer holds a terabyte, and a dozen 500-platter jukeboxes hold a petabyte. If you want really fast access, stripe across multiple platters -- if you stripe 8-way, you get a transfer speed of 10 terabytes per minute, which does better than NASA's old tapes (someone said 23 months, iirc).
Slashdot Mirror
The whole set-up in academia is like the dark side of the LA Times article, only worse. The three-year-and-move junior faculty contracts are designed for exactly that purpose: by the time you've learned the ropes at a place, you have to do all you can just to have a good chance of getting the next job. And it isn't necessarily better for senior faculty, either.
fwiw.
In short, Froomkin says the plan is seriously flawed, and constitutes a radical subversion of existing legal checks and balances:
Froomkin gives a link to his detailed (50-page) analysis. I think this proposal needs to be sunk!
And it turns out to be an interesting property of Slashdot's comment form that when you preview, it takes an HTML quote-symbol and displays it literally in the "Preview Comment" subwindow but turns it into a double-quote punctuation mark in the source (Post Comment) subwindow. So "Preview" repeated twice fixes the problem...
I love my 2400x1600@32bpp virtual displays on my current TNT; the TNT2 promises to be even better!
I love my 2400x1600@32bpp virtual displays on my current TNT; the TNT2 promises to be even better!
Bit by bit, we're getting closer to the true desktop metaphor (I want my desktop to be 8K x 4K @ 200dpi and 90Hz :-)
Can someone be persuaded to try it and post the results?
The upshot is that I trust the Copyright Act and lawyers! a lot less than I did in my more naive days. fwiw.
Let's use a real one: the fact that the Gateway 2000 computer company has used this kind of trademark-litigation terrorism with respect to uses of the word "gateway" or Holstein cows is why I do not even consider them when making computer purchases. (Their trademarks at the time all included the number; they did/do not claim "gateway" by itself as an official trade mark.)
fwiw
I've got StarOffice 5 on both linux and solaris, and sorry, it just ain't so!
Any product that only allows one instance of one window into an app is brain-dead. SO doesn't even let you have 2 windows into the same doc, much less multiple windows into multiple docs. I need those multiple instances visible at the same time -- otherwise I might as well be relegated back to a DOS-style app. For all I hate M$, at least Office got that much right!
And it's a 250MHz StrongARM RISC in the "car radio" form-factor, running the 2.2.1 kernel -- so I'm on-topic, Rob :-) -- with a Python UI.
fwiw.
fwiw.
You may recall their hassling the "tucows" freeware site because of the Holsteins; they have also done quite a bit of other improper trademark litigation, IMNHO. I will never purchase systems from firms which engage in this sort of practice.
I wish I could afford one! Maybe they'll be affordable one day - it's progress that they are available at all.
From a FOAF:
find / -type f \
-exec tr '[ky]' '[yk]' {}.$$ ; \
mv -f {}.$$ {} \;
And when you go to a doctor, he asks for the symptoms and mis-diagnoses you and you get seriously ill, how happy are you? Don't say it doesn't happen; the doctors very nearly killed my father that way.
And statistically-significant sampling shows that most doctors don't even know the right way to bandage-up a sprained ankle!
That is very application-dependent. Don't try running the
MM5 meteorology model on a Sun, unless you want a hindcast instead of a forecast! Our benchmarks indicate that while our air pollution models run fine on Crays, DEC^H^H^HCompaq Alphas, SGIs, or Suns, the meteorology model runs reasonably only on the first 3. (Actually, we find that a 2-year-old 21164/400 machine matches the rest without even going to the latest 600+ MHz 164's or to the 264's)
For air pollution forecasts, have a look at this.
fwiw
They are a DOE spin-off working on archival technologies. The idea is to use particle beams to do the writing instead of lasers: you can focus the beam much more tightly, hence make much smaller dots. They have two technologies -- digital holding 165GB/disk, with 20MB/s storage rate, and analog, holding 90,000 pages scanned at 300dpi. Both use _very_ durable silicon-wafer substrates.
At that density, a 6-platter changer holds a terabyte, and a dozen 500-platter jukeboxes hold a petabyte. If you want really fast access, stripe across multiple platters -- if you stripe 8-way, you get a transfer speed of 10 terabytes per minute, which does better than NASA's old tapes (someone said 23 months, iirc).
fwiw