Those of you using IE will need to switch to Mozilla. Those of you using Mozilla won't even notice the part that doesn't work under IE, it feels so natural.
Cool effect that works only under Mozilla and just feels right. Now who's at the disadvantage?
Thanks, Open Source Community, for helping us improve our document. Please keep sending us patches for it, we will certainly consider them.
(As you know, not many managers read Slashdot, or have much respect for the "Slashdot crowd". No one will listen to your shrill cries of "This is just FUD!" [In fact, I would hazard a guess that there is a fair number of managers who don't even know what the acronym "FUD" stands for.])
Calling/. "advocacy" is just an excuse incoherent, puerile screeds against MS, xxAA, broadband companies and whoever else we don't like because they won't give us their stuff for free.
You say/. has never pretended to be anything else. How about "News for Nerds, Stuff that Matters" (as if, but that's a different topic). Nothing about advocacy. (I interpret "Stuff that Matters" to be a modification of the "News" part, not an independent clause.)
I remember the advocacy newsgroups (particularly os/2). Lowest signal-to-noise ratio of all comp.os.*.* groups. You want to tag Slashdot w/that? No thanks.
Kudos to your users for even trying, by the way. A serious effort, failed or not, has got to be a useful datapoint. I assume you guys filed lots of bug reports.
Being in the kernel isn't magic
on
Mozilla RC3 Released
·
· Score: 2, Informative
People make this fallacious assumption a lot. Sure, you get more privilege, and you don't have to worry about timeslicing and competing w/other user processes, but there ARE other kernel "things" going on you have to compete with, and you still have to service device interrupts, in a minimal sort of way.
Plus, you can still do stupid things with locks and totally throw away your advantage.
Case in point: I myself did a benchmark of Samba on a vendor's Unix vs. SMB on NT 4.0 SP3 a few years back (on the same hardware, duh). (It wasn't quite formal enough to publish, we didn't get written permission from ZD Labs, etc. etc., you know the drill, but I did spend a month on it.) We outperformed NT by a factor of 2 under fairly high loads (~30 workstations on 3 sublans hitting the server as hard as they could). How could that be, when SMB was a kernel process on NT, but Samba was a user process on our Unix? My only explanation (apart from the possibility that I fucked up the benchmark [pretty small, I had some assistance from our local gurus]) is that NT/SMB do some stupid locking things in the kernel and slow themselves down.
I've seen a bunch of these over the years. "Americans are the stupidest bunch on the planet" etc. "Americans can't do math, oh my God, look at all the Indian engineers." This is an appeal by Americans to Americans, based on nationalism. "We can't fall behind other nations in math." Piffle.
Let's try this on for size: another study a few years ago (can't cite, it, sorry) revealed that even the Americans we consider "smart" (i.e., knowledgeable in math) have a surprising tendency not to believe that evolution is real (something like 30% of engineers). What's up with that? I wonder how many "foreigners", while being able to solve a partial differential equation at the drop of a hat, still believe in other goofy theories, like the usefulness of some totally bunk folk remedies. I personally know someone who might possibly be able to out-math me, but who believes that people of her ethnicity have such different biologies that Western medicine simply doesn't apply. What is up with that??
On the other hand, my son's school (here in America) has had at least one series of classes on critical thinking (e.g., seeing through propaganda and stereotyping). Not mathematical, but, in my opinion, way more useful than memorizing a bunch of antiderivatives (although, I admit, those are more immediately useful in a career).
So, as an American citizen, I'm slightly offended and embarrassed that the NSF is using nationalism to argue for more spending (essentially) on science-related fields. I certainly wish Americans (and the rest of world; are you listening to me, you people who voted for Milosevic?) were smarter, but I'm not a fan of using scare tactics and misleading arguments to accomplish this.
I gotta jump in here, being a Davidson alum (of yore, unfortunately). They're pretty serious about the honor code, and for the most part, I think everybody (including the "poor, oppressed" students) buy into that. The arguments I've read so far on this, that strict honor codes just lead to more rule-breaking, and honor codes should be flexible, are essentially just moral relativism. People sign statements when they accept an offer from Davidson that they are willing to live by the honor code, and they do so in the knowledge that the honor code has teeth.
The result is that people (when I was there, in the early 80's anyway) would leave expensive calculators lying around in the library w/out a moment's hesitation. Doors were hardly ever locked. And you took your final exam whenever you wanted, in whatever room of the main building you wanted. I can't remember if we had proctors (my memory's been fading ever since the blizzard of '89 -- or was that '98?), but even if we did, they wouldn't know which exams were open-book.
Anyway, it was great. And I still believe in a fairly painful honesty, and I am sorry sorry sorry the rest of the world doesn't. Arguing that the rest of the world is evil, so we should be too, is just wrong.
I understand the majority of security breaks are perpetrated by people who are already authorized to use the software in question, but do so in a way not consistent with their job description, as if the only security was the job description.
Insiders, in other words.
I went looking for security books a while ago and found lots of books on how to secure your web server and keep your machine from getting rooted and how to encrypt data sent over the wire. Now I see a book with a chapter on buffer overruns.
How hard can it be to write a book on keeping the various legitimate users of the system from snooping around the parts of the system they aren't supposed to go to? Doesn't anybody remember the scandal at the dept of Health & Human Services (I think that's the right dept.) in which users of the system were essentially surfing people's social security data for entertainment? I don't believe that had anything to do with buffer overruns or rooting servers or any of that hacker shit.
Wow. I always wonder about figures like that. I wonder how much hallway conversations, the superbowl and Christmas cost companies. Maybe it would be more cost-effective to eliminate those.
24 inches of shelf space devoted to deciphering those amazing IBM diagnostic codes (and other signs of thought put into how the user's going to cope when things wander off the main path).
I always used to sort of sneer at "undecipherable diagnostic codes" and the necessity to look them up. Now I long for the days of sufficiently detailed (RELEVANTLY detailed) diagnostics that I could understand and solve the problem without further futzing around.
I *knew* that story arc was self-destructive. I really enjoyed the "quirky" episodes a lot more (like Flukey and the Fiji Mermaid episodes). There's a better way to allude to vast mysteries than an alien conspiracy. And government cover-up conspiracies are a more-boring plot device than sheer government inertia and apathy, IMO.
How is that when MS says "It's cross-platform: it runs on Win98 AND Win2000" we all snicker, but when somebody says "It's cross-platform: it runs on all flavors of Unix" we don't even blink?
To me, true cross-platformness includes the ability to run on AS/400s, S/390s and VMS. Like emacs or slickedit or... perl?
Ha! I love this. Go with Microsoft, and you're on some sort of "PC economics model" treadmill. You pay what they want, when they want. Go with Linux and all you pay for is a sysadmin or two (or n). Very predictable costing: n persons' salaries, every year; no surprises.
John.
Debuggers can't back up, can't save state
on
Java IDEs?
·
· Score: 1
When I use a symbolic debugger, I invariably stop paying attention as I'm hitting next next next and suddenly I find the bug (duh) but I really wanted to stop at the line just before everything went to hell. Or for that matter, I stupidly single step past some crucial line where things started to go wrong. And I can't back up. This is caused by sheer monotony of hitting "next" so many times.
Also, in lieu of printfs, I've set up zillions of watchpoints and breakpoints-with-actions that print out values of variables and now it's time to quit the debugging session. Save state? Good luck! Assuming the debugger even claims to be capable of saving state, when you come back in to the debugger and attempt to reload that state (assuming that even works), what used to be line 584 is now line 592 and all your actionpoints are off (and useless). If I had done it with printfs and "if" stmts, the line number changes wouldn't be screwing me up.
Debuggers are great for getting stack traces and analyzing core dumps and looking at program state at the time of failure, but after that, it's back to studying code and inserting judicious printfs.
(I'm a former DGer.) I was under the impression that EMC bought DG exactly so they could get their hands on that "moldy old DG crap". Certainly the Aviion line wasn't that attractive. If the EMC salespeople don't want to sell a purchased company's products, that might be a post-acquisition integration (or higher-comission) issue rather than any reflection on the quality of the product.
Plus, I think it's still under active development, so "moldy old" doesn't really apply, either.
My current obsession. The problem-solving part is untangling routing messes. Doesn't require too much literacy, but does require the ability to recognize numbers (I need $285K to put this rail down and I've only got $240K; plus I'll need to spend $50K on an engine.) And it's business-oriented (if you consider robber barons to be business-oriented), so you can get 'em into that capitalist/happy-contributor-to-society mindset.
John.
"Finnish programing whiz Linus Torvalds"
on
NYSE Goes To Linux
·
· Score: 0, Offtopic
Numbers of large companies and universities over the last two years have shifted their computer systems to the software developed by Finnish programing whiz Linus Torvalds.
Interesting how the popular media (or the popular business media, or maybe just this magazine) still can't bring itself to say that Linux is developed by a horde of developers, only loosely organized and working under a model that doesn't quite fit the capitalist picture.
They always have to have some identifiable person or company to point to. Just another version of "who do you sue?".
Slashdot Mirror
Those of you using IE will need to switch to Mozilla. Those of you using Mozilla won't even notice the part that doesn't work under IE, it feels so natural.
Cool effect that works only under Mozilla and just feels right. Now who's at the disadvantage?
When was CDW determined to be a monopoly?
Thanks, Open Source Community, for helping us improve our document. Please keep sending us patches for it, we will certainly consider them.
(As you know, not many managers read Slashdot, or have much respect for the "Slashdot crowd". No one will listen to your shrill cries of "This is just FUD!" [In fact, I would hazard a guess that there is a fair number of managers who don't even know what the acronym "FUD" stands for.])
Calling /. "advocacy" is just an excuse incoherent, puerile screeds against MS, xxAA, broadband companies and whoever else we don't like because they won't give us their stuff for free.
/. has never pretended to be anything else. How about "News for Nerds, Stuff that Matters" (as if, but that's a different topic). Nothing about advocacy. (I interpret "Stuff that Matters" to be a modification of the "News" part, not an independent clause.)
You say
I remember the advocacy newsgroups (particularly os/2). Lowest signal-to-noise ratio of all comp.os.*.* groups. You want to tag Slashdot w/that? No thanks.
...and in which it turns out several of the warriors' leaders are half-trolls themselves.
Every good story needs a plot twist or two.
Could you quote me chapter and verse on that? Maybe this is a way to get me reading the bible.
(I've seen this assertion a lot, but have usually found the "pornographic" parts of the bible less lascivious than other stuff I've read.)
Kudos to your users for even trying, by the way. A serious effort, failed or not, has got to be a useful datapoint. I assume you guys filed lots of bug reports.
Plus, you can still do stupid things with locks and totally throw away your advantage.
Case in point: I myself did a benchmark of Samba on a vendor's Unix vs. SMB on NT 4.0 SP3 a few years back (on the same hardware, duh). (It wasn't quite formal enough to publish, we didn't get written permission from ZD Labs, etc. etc., you know the drill, but I did spend a month on it.) We outperformed NT by a factor of 2 under fairly high loads (~30 workstations on 3 sublans hitting the server as hard as they could). How could that be, when SMB was a kernel process on NT, but Samba was a user process on our Unix? My only explanation (apart from the possibility that I fucked up the benchmark [pretty small, I had some assistance from our local gurus]) is that NT/SMB do some stupid locking things in the kernel and slow themselves down.
So, the kernel ain't magic.
John.
...Amazon is evil. (Except maybe you.)
I've seen a bunch of these over the years. "Americans are the stupidest bunch on the planet" etc. "Americans can't do math, oh my God, look at all the Indian engineers." This is an appeal by Americans to Americans, based on nationalism. "We can't fall behind other nations in math." Piffle.
Let's try this on for size: another study a few years ago (can't cite, it, sorry) revealed that even the Americans we consider "smart" (i.e., knowledgeable in math) have a surprising tendency not to believe that evolution is real (something like 30% of engineers). What's up with that? I wonder how many "foreigners", while being able to solve a partial differential equation at the drop of a hat, still believe in other goofy theories, like the usefulness of some totally bunk folk remedies. I personally know someone who might possibly be able to out-math me, but who believes that people of her ethnicity have such different biologies that Western medicine simply doesn't apply. What is up with that??
On the other hand, my son's school (here in America) has had at least one series of classes on critical thinking (e.g., seeing through propaganda and stereotyping). Not mathematical, but, in my opinion, way more useful than memorizing a bunch of antiderivatives (although, I admit, those are more immediately useful in a career).
So, as an American citizen, I'm slightly offended and embarrassed that the NSF is using nationalism to argue for more spending (essentially) on science-related fields. I certainly wish Americans (and the rest of world; are you listening to me, you people who voted for Milosevic?) were smarter, but I'm not a fan of using scare tactics and misleading arguments to accomplish this.
</rant>
John.
I gotta jump in here, being a Davidson alum (of yore, unfortunately). They're pretty serious about the honor code, and for the most part, I think everybody (including the "poor, oppressed" students) buy into that. The arguments I've read so far on this, that strict honor codes just lead to more rule-breaking, and honor codes should be flexible, are essentially just moral relativism. People sign statements when they accept an offer from Davidson that they are willing to live by the honor code, and they do so in the knowledge that the honor code has teeth.
The result is that people (when I was there, in the early 80's anyway) would leave expensive calculators lying around in the library w/out a moment's hesitation. Doors were hardly ever locked. And you took your final exam whenever you wanted, in whatever room of the main building you wanted. I can't remember if we had proctors (my memory's been fading ever since the blizzard of '89 -- or was that '98?), but even if we did, they wouldn't know which exams were open-book.
Anyway, it was great. And I still believe in a fairly painful honesty, and I am sorry sorry sorry the rest of the world doesn't. Arguing that the rest of the world is evil, so we should be too, is just wrong.
John.
Sounds like a goofy question, I know.
I understand the majority of security breaks are perpetrated by people who are already authorized to use the software in question, but do so in a way not consistent with their job description, as if the only security was the job description.
Insiders, in other words.
I went looking for security books a while ago and found lots of books on how to secure your web server and keep your machine from getting rooted and how to encrypt data sent over the wire. Now I see a book with a chapter on buffer overruns.
How hard can it be to write a book on keeping the various legitimate users of the system from snooping around the parts of the system they aren't supposed to go to? Doesn't anybody remember the scandal at the dept of Health & Human Services (I think that's the right dept.) in which users of the system were essentially surfing people's social security data for entertainment? I don't believe that had anything to do with buffer overruns or rooting servers or any of that hacker shit.
I do, I do! But I agree w/you, ESR is a total gadfly.
Wow. I always wonder about figures like that. I wonder how much hallway conversations, the superbowl and Christmas cost companies. Maybe it would be more cost-effective to eliminate those.
"I feel your pain."
:)
Well, at least Clinton left *something* behind.
Maybe I'm clueless, but I think the time is ripe for client-side Java to return, now that the plugin has matured.
How about a book on client-side enterprise Java using the plugin (applets) or Web Start?
Or a book on Struts (Jakarta)?
Or a book on well-engineered Javascript (as opposed to cute hacks)?
Or a book on delivering well-engineered web apps (objects/widgets that express themselves in html + javascript)?
John.
24 inches of shelf space devoted to deciphering those amazing IBM diagnostic codes (and other signs of thought put into how the user's going to cope when things wander off the main path).
I always used to sort of sneer at "undecipherable diagnostic codes" and the necessity to look them up. Now I long for the days of sufficiently detailed (RELEVANTLY detailed) diagnostics that I could understand and solve the problem without further futzing around.
I *knew* that story arc was self-destructive. I really enjoyed the "quirky" episodes a lot more (like Flukey and the Fiji Mermaid episodes). There's a better way to allude to vast mysteries than an alien conspiracy. And government cover-up conspiracies are a more-boring plot device than sheer government inertia and apathy, IMO.
How is that when MS says "It's cross-platform: it runs on Win98 AND Win2000" we all snicker, but when somebody says "It's cross-platform: it runs on all flavors of Unix" we don't even blink?
To me, true cross-platformness includes the ability to run on AS/400s, S/390s and VMS. Like emacs or slickedit or... perl?
John.
Ha! I love this. Go with Microsoft, and you're on some sort of "PC economics model" treadmill. You pay what they want, when they want. Go with Linux and all you pay for is a sysadmin or two (or n). Very predictable costing: n persons' salaries, every year; no surprises.
John.
Also, in lieu of printfs, I've set up zillions of watchpoints and breakpoints-with-actions that print out values of variables and now it's time to quit the debugging session. Save state? Good luck! Assuming the debugger even claims to be capable of saving state, when you come back in to the debugger and attempt to reload that state (assuming that even works), what used to be line 584 is now line 592 and all your actionpoints are off (and useless). If I had done it with printfs and "if" stmts, the line number changes wouldn't be screwing me up.
Debuggers are great for getting stack traces and analyzing core dumps and looking at program state at the time of failure, but after that, it's back to studying code and inserting judicious printfs.
John.
(I'm a former DGer.) I was under the impression that EMC bought DG exactly so they could get their hands on that "moldy old DG crap". Certainly the Aviion line wasn't that attractive. If the EMC salespeople don't want to sell a purchased company's products, that might be a post-acquisition integration (or higher-comission) issue rather than any reflection on the quality of the product.
Plus, I think it's still under active development, so "moldy old" doesn't really apply, either.
John
ok, ok, i'll write some sort of body to get around the lameness filter
not sure I need to expand on the subject line, though.
My current obsession. The problem-solving part is untangling routing messes. Doesn't require too much literacy, but does require the ability to recognize numbers (I need $285K to put this rail down and I've only got $240K; plus I'll need to spend $50K on an engine.) And it's business-oriented (if you consider robber barons to be business-oriented), so you can get 'em into that capitalist/happy-contributor-to-society mindset.
John.
Interesting how the popular media (or the popular business media, or maybe just this magazine) still can't bring itself to say that Linux is developed by a horde of developers, only loosely organized and working under a model that doesn't quite fit the capitalist picture.
They always have to have some identifiable person or company to point to. Just another version of "who do you sue?".
John.