Slashdot won't accept the post I wrote in reply to yours. So I put it on my web server here where there is no "Lameness Filter" to get in the way. To reply to that, come back here (if you can).
Maybe some day they will make the Lameness Filter explain what it thinks is lame so we know what to fix.
I don't want my windows re-arranged. Changing the desktop size can't really be done without that. So for me, panning a smaller video view over a larger desktop is right. As far as the WM knows, the desktop size is what it works with and I want that to stay the same (usually). I don't want the X server re-arranging windows if they need to be re-arranged; I want the WM to do that (so I have pluggable choices on how that's done). Different people like different things, and as much of that choice should be where choices are most easily handled... in the WM.
This is too utterly obvious to be innovative. This patent should be used as the "poster boy" of reforming the patent system. I bet those guys at Bay Networks thought they were so clever in doing what thousands of others would consider to be quite obvious (but had other important work to do, so they never implemented it). That would tend to make me think that the level of IQ among engineers at Bay Networks is not very high if this kind of thing is something they consider to be an invention. Now we know the examiners in the USPTO are low IQ. But now I have to add in the engineers at Bay Networks, too. Boycott time.
I'm only talking about how the software does it. One of the intents of the refresh is to make sure some stray pixels that failed to get cleared out are in fact clear. That's done by literally writing some blank color, usually the current background color or background image, then redrawing or rerendering every object on the screen. Unfortunately programmers thought refresh() meant rerender() because the latter was being done by the former (after the blanking).
Now days, there are more tools that don't do this, as well as much faster transfers to video bus, and accelerators doing the hard work even faster, so fortunately, the flashing is now more rare. But there are some things still around that do it. For example, when I minimize then maximize the Acrobat Reader from Adobe, while on a PDF page that is more complex with constructed images, it redraws the whole construction all the way to my window. Sometimes it takes several seconds for it to finish.
Back in the DOS days, when I wrote my graphical tool, almost everything did this blanking thing. I was using another tool for a while, but I found that the only way to ensure the screen was up to day was to call refresh(), and it blanked the screen out first, then wrote all the windows onto the screen from bottom first to top last. Whatever was not in some window was left in the blanked out state. My tool did it a different way. As the window structures were built, it maintained a cache of pointers (and these were for characters, not pixels, as this was before Windows 3.X even came out, so most things were text based) for every character (not the most efficient way, but it was plenty fast enough on the 8 to 16 MHz machines of the day). When a window position or layout change was made, the cache was rebuilt. When a refresh call was made, it went through the cache in raster scan order, picked up a pointer, grabbed the character pointed to, and put that character in the corresponding position in the video buffer. It was still ightning fast and absolutely smooth. There was no flashing or flicker anywhere. The hardware did its part (it didn't have memory bus collisions between video scan and buffer writes, as some earlier video cards had), and the software did its part.
If you want to have windows re-arrange, then I guess you and I are looking for different features. What you are looking for is a better window manager. It would have to make sure X can notify the WM of the desktop size change, and possibly even allow the WM to do the change itself (so you can have a cute pop-up window/menu thingy to do it).
2. So you can do it, SO FUCKING WHAT! It needs to be part of standard distros' scripts otherwise the common man is not helped. (just because you can do it, doesn't mean someone else can)
It can be done. Take the issue up with the people who prepare your distribution. Or are you suggesting I get into making my own distribution?
3. Force the vendors? Believe it or not you are not entitled to FORCE anyone to do anything. Yes we need standards, though they can't be forced. They need to be accepted. (you should never go for a job in pr)
Ever heard of boycotts to force companies to do things better? It happens. That's what I was referring to... consumer motivations.
4. I'll try hard not to invalidate my point 2. Once you can do things it's easy, what we need is intuative operations (and not intuative used as a pr tool as it often is!)
Too often, lots of people, like those who build distributions, spend more time trying to keep doing things the same old way. Well, where the old way works, that's great. Where it doesn't, they need to change. Too often they don't. This is mostly a distribution building problem.
5. You don;t seem to understand the prtoblem atall. Erasing is NOT the problem, well not unless we have THICK AS PIGSHIT programmers, it's more a we do it but can't do it all in 1 frame so it looks shit, smoothnesss is presented by drawing to a off screen buffer and putting onto the screen when it's ready. this may be slow on slow machines but it never flickers.
If you slow things down extreme, you'll see the flashing really is a screen being updated, blanked out, and updated again. One thing that happens is when a lot of functions are finishing up, they refresh. Over the course of some updates, this may happen several times. That's not bad if the refresh doesn't blank everything to background color and re-draw. It really is a blanking operation very often (not only on X Windows, but also on MS Windows).
6. If programmers knew how to deal with the real world, they'd stop programming and go out into it. I fail to see how this sloves the problem;-)
OK, maybe spanking will:-)
7. Don;t bne paranoid, they are after you.
8. Again, passing the blame will get you nowhere, except posibly higher up the corperate ladder.
Well, I have to agree with the part about passing the blame gets you up the corporate ladder. But the blame for the extreme number of drivers to deal with, and the problem of having to thus deal with so many drivers and modules, does belong to the hardware designers.
9. HELLO! Gte a clue, soft wrapping it a good thing, it enables me to read my email formated to my screen resolution and not at your, etc.
I understand what that is, but you need some form of specification to say what is or is not in a group to be formatted, and how. ASCII doesn't have this capability, so we shouldn't expect it to magically happen with ASCII. HTML does, as do many other text formats. Try one. But if after you edit it, you send it as ASCII, be sure the newlines are in there, or else send it in a format that can be automatically formatted correctly, such as HTML (yes, I do HTML).
10. Yes but that needs SETTING UP, as does X. WE need better autodetection, I mean I don't want to have to tell X what resolutions i would like to change bettween before i do. Plus i think he more wanted a nice control panel style changer, that ONLY let's you do valid resolutions automatically.
The autodetection problem gets back to the extreme diversity of hardware interfacing which requires so many drivers and conflicting probes that can hang systems. If the hardware makers were to adopt certain interfaces that allow the low bandwidth operations (e.g. set video mode) to be done in the same form as everyone else, a lot of problems fall. But if you want to see it done in software, working for the full range of existing hardware, expect it to be a huge project. Dive in and try it yourself.
And there aren't just a few select resolutions that are valid. There can be hundreds or thousands of possible resolutions. Of course only a few are needed for *normal* people to choose from.
Most of these problems come down to either the developers of some application doing things better, or the distribution preparers doing things better. I doubt we will solve anything in Slashdot. We have to focus on each individual case. And don't forget that most of the development is by programmers, for programmers. If we want to have it become for non-programmers, we do have to start thinking different. I sure hope that doing that doesn't break the power tools programmers use.
Free speech is about free speech. The <b>first amendment</b> is about limiting the government's ability to infringe on free speech.
Do you really believe that only the government has a moral duty to consider how their actions affect free speech? Yes, the government is the one with the <b>legal</b> duty to not infringe on free speech, but free speech is more than just the law, it's a good idea!
So far I agree with you.
Yes, we <i>can</i> boycott who we chose, including the innocent. But should we? Is it a desireable state? We're replacing the old means of speech with a new one, one that relies entirely on private property. How should we design it? Does designing it to run entirely on private property mean that we want private property rights to trump the old rights the founders of the USA felt fit to write into the constitution?
I don't see private property being the issue. Speech has always been either based on private property (buy a printing press and start publishing) or public property (stand in the town square and start your speech and see if anyone has any interest). We're not designing anything to be private property; it's either private or it's public.
These are hard questions, not necessarily rhetorical, since clearly some writers here do think that now that we've escaped the bounds of government we should ignore the principles it was designed with in free states.
My right to free speech and my right to ignore anyone are equal rights to me. The only reason free speech is such an issue is because governments (and even corporations) so often try to suppress that right. That's been so extreme even hundreds of years ago that some people setting up a new government in the late 1700's decided to specifically address the issue that was a major problem. The right to ignore hasn't been a problem, but that doesn't make it any less of a right.
There are other ways to punish bad ISPs without blocking the mail of innocent users. In particular, the obvious one rarely implemented is not to block mail from sites but to throttle its volume down, so that individual mails go through and bulk mails are blocked. A coordinated effort to do this, as I have written about on my web site, would cause vastly less collateral damage.
While the spammer at an ISP is spewing out millions, I get one or a few. Throttling them has no effect. I still get hit by hundreds of different IP addresses. If the realm of throttling is just one IP at a time, it won't work. If it covers all the IPs as a group, it has adverse affect on everyone.
And worse, this idea removes the incentive, because then the ISP won't be losing customers. There has to be "collateral damage" to make the boycott have an impact on an ISP. Maybe this is surprising to you, but lots of ISPs make the decisions about how they do things based on what will yield them the greatest revenues and profits. If they don't feel that their customers will leave because of some action, then they will have no reason not to do so if they feel it will bring in some revenue. And that action might be to accept a known spammer as a customer.
Do we have a duty to not cause collateral damage? I think so.
What other mechanism of causing customers to leave an ISP is there besides "collateral damage"? Spammers have a duty to stop spamming. Let me know when they stop so I can resume listening to free speech.
So if one browser gets better, and then because of the pressure another gets better, too, this is bad? Maybe we should remove some features from one of them to make another look better? It's sad that we would have to downgrade the capability of something before we are able to make a choice.
2. Prompting for a filesystem scan.
I rewrote my rc/boot scripts myself from scratch. I haven't had this problem for 3 years.
3. Printing needs to be easier to configure.
We need better standards among printers. Much of the problem is due to so many different kinds of printers, different drivers, different data formats. One single standard is needed and vendors must be force to comply.
4. Make it easy for the user to find out how to do things.
I still haven't figured out how to do a number of things on my MS Windows 98 machine. For example, how do I tell Windows that my hardware clock runs as UTC and that it should still show me my local time.
5. Cleaner redraws.
This is more of a programming problem. Certain programmers think that they need to first erase the screen then rewrite it. Back before Linux, I wrote an editor for DOS, and I wrote my own screen window manager for it. The editor could simply open up window objects and update them much like curses, but simpler. When refresh was called, the screen was updated, but there was no flicker because it was never erased first. It simply updated everything, period. Parts that were changing content just changed. Parts that were not changing, didn't. And mine was so fast I could still do scrolling by full screen rewrites even on a 16 MHz machine.
6. Die stray processes, die!
Programming problem again. Teach programmers how to deal with the real world.
7. Easy way of sharing files.
Why do you want access to my files? Leave me alone.
8. Sound support.
Like printers, this is a vendor problem. Find vendors who do a better job of not always changing the driver-to-hardware interface, and favor them over the vendors that keep screwing people over with the next board version. There is no reason every piece of hardware needs to have its own driver disk included, even for MS Windows (and this is a big cause of many system problems in Windows, too... Bill Gates has said so).
9. No common editor which supports "soft wrapping."
What you are asking for is to show text as if it had newlines, when in fact it has none. Maybe you should be writing HTML instead of plain ASCII text. Don't mail it to me w/o newlines. But if you want to be able to reformat a range of text, maybe you should try emacs.
10. No easy way to configure X - especially change resolution on the fly.
I just changed my resolution on the fly while entering this line of text by pressing the Ctrl-Alt-KeypadMinus combo. Then I pressed Ctrl-Alt-KeypadPlus to revert back.
If you don't boycott the bad ISPs... the ones that keep signing up spammers who leave their previous ISP so they can have new address space not yet listed... then spammers will constantly be showing up at new addresses, and listing them does little good. You have to boycott bad ISPs (the ones who harbor spammers) to make them either turn good, or move the good customers off to a better ISP. Of course there will be "collateral damage"... but not to the ISPs that respond appropriately and cut off spammers. The non-spamming customers can complain to the ISP (this is part of the motive) or just move to another ISP.
Being listed in a blacklist is not the same as being indicted for a crime. It's a boycott. Different rules. Different standards. If you think that SPEWS, for example, has more collateral damage than you want your mail server to participate in, then don't use it.
Free speech is not about making sure everyone gets to say whatever they want to whoever they want however they want. It's about making sure government doesn't step in and limit it. Free speech cannot be used to justify stealing my printing press. Indeed, such a theft would deprive me of my right to free speech in such a case. Likewise, you have no right to steal my network server time for your own speech. I own it for mine (or that of my customers who pay me for its use). This is why it's a property issue. It's no tougher for me to deal with theft of my server time than dealing with theft of my printing press.
If you are communicating with me about matters we have a relationship based common interest in, then that is expected and fine. For example if you were to e-mail your reply to this post, that is reasonable to expect because by posting this (even if it wasn't in response to yours), there is an implied expectation of a response. Trying to sell me penis enlargers, golf balls, ink cartridges, and mortgages is not applicable communication, unless I asked (in some form) where I could find those things to buy.
If you really have written about better methods that don't generate false positives, then please show me, and please point out specific ways these methods work to drive incentives to ISPs to disconnect spammers and to not sign up at all known spammers. Spamming will not decline if ISPs have no counter incentive (to the revenues) to choose to not allow spammers in their network.
How else can you boycott the ISP w/o collateral damage? SPEWS does not list the ISP, and hence, no collateral damage, until the ISP has had plenty of time to cut off the spammer. In order to increase the level of pressure on the ISP, more of their address space has to be listed to "encourage" them to cut off the spammer. The usual first listing is the whole/24 the spammer is in (if they weren't doing it from the whole/24 in the first place). Maybe they will start listening once their own customers complain (and that's the proper place for the customers to complain to, their ISP). If they continue to ignore the problem, then eventually the whole ISP will be listed. If it's a multi-level ISP, their upstream starts to get listed, too.
The philosophy SPEWS appears to be using, and one I now agree with (previously I did not, but sometimes my opinions do change... hey, I'm open minded), is that the spam problem will not go away by blocking only the spammers. ISPs have to play a part by not signing up known spammers, and cutting off spammers that got signed up because they were not known at first. Blocking spammers alone will be a never-ending battle because then there is no incentive for any ISP to turn them away and they just keep moving around to evade the blocking. To end spamming, the ISPs have to quit offering them services, or we have to quit accepting traffic from the set of ISPs that do harbor spammers.
It looks like collateral damage, but it's just another form of boycott. If I organize a boycott against my local newspaper, then the advertisers suffer because fewer people read their ads. And such boycotts are known to even extend to boycotting the advertisers if things get bad (and spam right has gotten very bad already). Is that fair to the advertisers? Of course not. But that's the nature of the activity; it is, among other things, trying to encourage the advertisers to cease advertising there. So in the same way, by boycotting a whole ISP address space, the idea is to encourage their customers to change to another ISP, until the ISP changes their ways.
Palestinians are not saying that Israel has no right to exist. What they are saying is that Israel has no right to taking Palestinian land (all 100% of it) as the basis for their existance.
What is needed to stop this is a moderating system which ranks the various traded products, as identified by their MD5 checksum signatures, according to some "measure of quality". By rank ordering, it cannot be used to entirely shutdown a trading network since everything would still be available. Products at 50 out of 100 would have received a ratio of good vs. bad moderations better than 50% of other products, and worse than the other 50% of products. It would not necessarily be a 50/50 good/bad moderation. Thus flooding of bad moderations across the board would have no effect, though it could be used to drive very specific classes of products down the list. But eventually, people would see the abuse and mod them back up. It would be sort of like moderation on slashdot, but everyone gets to play.
Now would it be possible to have selective moderation like slashdot has? Only a central authority could do that the way slashdot does. The big question would be judging who gets moderation points. As far as I know, on slashdot, it's almost entirely automated. With product trading, it would be harder to measure the quality by automation, so someone has to manually make the judgement calls and that brings some risks as well.
If individuals could be identified uniquely in some way, without the risk of exposing real identity, then meta moderation might work. One way to do that would be a slow rate of generating some kind of signed digital certificate that allows only so many to be generated at a time per network that receives it (and no personal identifying info included, and no records kept). Moderations and meta moderations would be signed by these anonymous certificates. You wouldn't know who moderated, but what you would know is that a group of moderations by the same certificate are probably from the same person and can be judged accordingly, good or bad. Excessive levels of moderation would also weaken your merit and derate your contributions.
PRINCETON, N.J., July 2/PRNewswire-FirstCall/ -- RCN Corporation (Nasdaq: RCNC - News) announced the launch of a new "super-charged" high-speed Internet service in its San Francisco and Los Angeles markets. Known as MegaModem(SM), it enables RCN's California customers to access the Internet at download speeds of up to 3 megabytes per second (Mbps), double the company's standard downstream speeds of up to 1.5 Mbps, and up to twice as fast as competing cable modem and Digital Subscriber Line (DSL) services.
I'm sure it's really megaBITS per second, otherwise the company would surely have promoted it as 24 megaBITS per second if the speed really was 3 megaBYTES per second (so they could use the larger number). You don't generally see 1.5 megabits per second promoted as 187.5 kilobytes per second. But this does show how reporters are still subject to making technical errors, which I suspect is due to their lack of knowledge of technical details. At least they got the case of Mbps correct.
Not all server environments can be just trivially made to be multithreaded and still work securely. Multithreading means more than just sharing the codebase. It means sharing the data and sharing the access permissions. It means a huge security exposure in servicing different users. So that is where the 1000 processes comes from (and its not a low estimate, either... I just talked to one person the other day who had as many as 12000 processes running... fortunately not JVMs... during a peak of usage). If Java is to get beyond lame stuff like shopping carts, it needs to become wise to more isolated secure environments, and still do it efficiently.
1000 processes doing just-in-time compiling? Now tell me how that resultant machine code is shared securely between processes. If compilation to native machine code can be done, why not do so before run time, and then the stored machine code image can have its one copy in RAM shared by the many processes running. This wouldn't have been an issue with applets in a browser, but it certainly is in a server.
I like the Java language... but this Java Virtual Machine is a hindrance. For things like downloaded applets that need to run in anyone's browser on any machine, a portable run-time is essential. But for building server-side applications, the JVM is the problem. So I'm looking for a compiler system with libraries that compiles Java to native machine code (with many platforms supported), and still does everything any Java program would expect... except for things like being able to share the runtime code, both the executeable and the libraries, across thousands of process instances, and the raw speed native code gets. Oh, and GPL, LGPL, or BSD style licensing would be nice.
Nearly 200 different non-existant usernames in my various domains are being spammed, many on a continuing basis. These are usernames that have never existed, and never accepted delivery of mail, so they are definitely not confirmed opt-in's for anything. It just shows how far spammers will go, and how they never clean up their lists (as if that would help real people).
If you're only going to pay $50k for techs, then you're going to get the ones who can't spell. You get what you pay for. If you don't need to have techs writing reports, then the $50k techs will probably do just fine for you. If you want a tech that can also do things like write proper reports to management, and document the systems and programs they put together, then I suggest you start paying more so you get a better pool of applicants and can choose the ones that fit your needs. Don't whine about the porr quality of techs if all you're offering to hire them is a lousy salary level.
Unfortunately, Slashdot doesn't pick and choose its readers and posters, so of course you get all kinds here. But there are many out there who can spell, and who can write documentation (some who can just don't like to, though).
Having the government ban the spam content might well be in violation of the 1st amendment. Banning fraud (e.g. using someone else's identity, or a fictional one) isn't. That won't cut out spam, but it can at least narrow the scope of what originates in the USA (or another country if they have a similar law).
Of cource spam makes money. But not for everyone. There aren't enough consumers available for the practice of spamming to reach into the thousands of people doing what Scelson does. At that point the net just grinds to a halt from overload (despite being designed to resist military attacks during war).
You are so utterly ignorant of the law. Maybe that's why you are keeping yourself anonymous. Wanna cite the federal law you think exists to force businesses to do business with just anyone? You can't, because it doesn't exist.
As for anti-spammers, there is no interference whatsoever. They are simply boycotting. They have a right to boycott any business they choose. Oh, in case you hadn't noticed, they do have the right to refuse to do business with whomever they don't want to. And if they don't want to do business (trade packets) with an ISP that keeps spammers connected, that's their choice and their right. Now the ISP decides whether they want to do business with the spam side of the net, or the non-spam side. Seems more and more are choosing the non-spam side.
By making custom machines, local integrators do not have to lock themselves into those mass quantity deals that they need in order to make sure their huge volume stays consistent. This lets them pick and choose what's cheaper today, while Dell and the other big computer makers are stuck taking shipments from their long term contracts. Only if there comes to be a parts shortage will the situation reverse.
As for the service part, the one thing you left out is that when the local integrator does send a tech out for on-site service, it's more often the very same guy who therefore knows more about the problem history the customer might be having, and won't start over from the beginning every time like you'd get if Dell sends a different sub-contracted tech out each time.
Also, if the problem is due to some incompatible piece of hardware, the local integrator is much more free to use some different brand or model of part to replace it and get things working, while the sub-contracted tech from Dell has to stick with the Dell brand products (so the next tech will understand it).
While the unemployment rate of the US population in general is a mere 6%, among techies, unemployment plus underemployment is somewhere between 25% and 35%. Techies are not that big a portion of the whole population. With no disposable cash, of course they won't spend where it can be avoided. Of course there will be many who steal music even if well employed, but many others won't.
Impossible is coding 100000 different values into a 16 bit field. That is where "cheap" isn't one of the choices as you have to expand the field... and that may mean recompiling and maybe even recoding everything else that expected to use just 16 bits.
Impossibility depends on how you state the problem. If the manager says "use a 16 bit field to code the 100000 values" then it is impossible. Tell them so and they might not believe you (in part because they may have come from a background where they are drilled with concepts like "nothing is impossible if you try hard enough", even though we know that is not true). Tell them 100000 requires at least 17 bits, and that its quicker to just use 32 bits, and then at least they haven't heard the word "impossible".
If you really do have to tell them it is impossible, maybe if you relate it to something they can understand, they can get the message. "That's like running a one minute mile". "A car can do that". "A car is a 32 bit field".
Slashdot Mirror
Actually, suits ask these three important questions:
Slashdot won't accept the post I wrote in reply to yours. So I put it on my web server here where there is no "Lameness Filter" to get in the way. To reply to that, come back here (if you can).
Maybe some day they will make the Lameness Filter explain what it thinks is lame so we know what to fix.
I don't want my windows re-arranged. Changing the desktop size can't really be done without that. So for me, panning a smaller video view over a larger desktop is right. As far as the WM knows, the desktop size is what it works with and I want that to stay the same (usually). I don't want the X server re-arranging windows if they need to be re-arranged; I want the WM to do that (so I have pluggable choices on how that's done). Different people like different things, and as much of that choice should be where choices are most easily handled ... in the WM.
This is too utterly obvious to be innovative. This patent should be used as the "poster boy" of reforming the patent system. I bet those guys at Bay Networks thought they were so clever in doing what thousands of others would consider to be quite obvious (but had other important work to do, so they never implemented it). That would tend to make me think that the level of IQ among engineers at Bay Networks is not very high if this kind of thing is something they consider to be an invention. Now we know the examiners in the USPTO are low IQ. But now I have to add in the engineers at Bay Networks, too. Boycott time.
I'm only talking about how the software does it. One of the intents of the refresh is to make sure some stray pixels that failed to get cleared out are in fact clear. That's done by literally writing some blank color, usually the current background color or background image, then redrawing or rerendering every object on the screen. Unfortunately programmers thought refresh() meant rerender() because the latter was being done by the former (after the blanking).
Now days, there are more tools that don't do this, as well as much faster transfers to video bus, and accelerators doing the hard work even faster, so fortunately, the flashing is now more rare. But there are some things still around that do it. For example, when I minimize then maximize the Acrobat Reader from Adobe, while on a PDF page that is more complex with constructed images, it redraws the whole construction all the way to my window. Sometimes it takes several seconds for it to finish.
Back in the DOS days, when I wrote my graphical tool, almost everything did this blanking thing. I was using another tool for a while, but I found that the only way to ensure the screen was up to day was to call refresh(), and it blanked the screen out first, then wrote all the windows onto the screen from bottom first to top last. Whatever was not in some window was left in the blanked out state. My tool did it a different way. As the window structures were built, it maintained a cache of pointers (and these were for characters, not pixels, as this was before Windows 3.X even came out, so most things were text based) for every character (not the most efficient way, but it was plenty fast enough on the 8 to 16 MHz machines of the day). When a window position or layout change was made, the cache was rebuilt. When a refresh call was made, it went through the cache in raster scan order, picked up a pointer, grabbed the character pointed to, and put that character in the corresponding position in the video buffer. It was still ightning fast and absolutely smooth. There was no flashing or flicker anywhere. The hardware did its part (it didn't have memory bus collisions between video scan and buffer writes, as some earlier video cards had), and the software did its part.
If you want to have windows re-arrange, then I guess you and I are looking for different features. What you are looking for is a better window manager. It would have to make sure X can notify the WM of the desktop size change, and possibly even allow the WM to do the change itself (so you can have a cute pop-up window/menu thingy to do it).
Only one installer question should be needed:
Any more than that, and half the world would not be able to install it right.It can be done. Take the issue up with the people who prepare your distribution. Or are you suggesting I get into making my own distribution?
Ever heard of boycotts to force companies to do things better? It happens. That's what I was referring to ... consumer motivations.
Too often, lots of people, like those who build distributions, spend more time trying to keep doing things the same old way. Well, where the old way works, that's great. Where it doesn't, they need to change. Too often they don't. This is mostly a distribution building problem.
If you slow things down extreme, you'll see the flashing really is a screen being updated, blanked out, and updated again. One thing that happens is when a lot of functions are finishing up, they refresh. Over the course of some updates, this may happen several times. That's not bad if the refresh doesn't blank everything to background color and re-draw. It really is a blanking operation very often (not only on X Windows, but also on MS Windows).
OK, maybe spanking will :-)
Well, I have to agree with the part about passing the blame gets you up the corporate ladder. But the blame for the extreme number of drivers to deal with, and the problem of having to thus deal with so many drivers and modules, does belong to the hardware designers.
I understand what that is, but you need some form of specification to say what is or is not in a group to be formatted, and how. ASCII doesn't have this capability, so we shouldn't expect it to magically happen with ASCII. HTML does, as do many other text formats. Try one. But if after you edit it, you send it as ASCII, be sure the newlines are in there, or else send it in a format that can be automatically formatted correctly, such as HTML (yes, I do HTML).
The autodetection problem gets back to the extreme diversity of hardware interfacing which requires so many drivers and conflicting probes that can hang systems. If the hardware makers were to adopt certain interfaces that allow the low bandwidth operations (e.g. set video mode) to be done in the same form as everyone else, a lot of problems fall. But if you want to see it done in software, working for the full range of existing hardware, expect it to be a huge project. Dive in and try it yourself.
And there aren't just a few select resolutions that are valid. There can be hundreds or thousands of possible resolutions. Of course only a few are needed for *normal* people to choose from.
Most of these problems come down to either the developers of some application doing things better, or the distribution preparers doing things better. I doubt we will solve anything in Slashdot. We have to focus on each individual case. And don't forget that most of the development is by programmers, for programmers. If we want to have it become for non-programmers, we do have to start thinking different. I sure hope that doing that doesn't break the power tools programmers use.
So far I agree with you.
I don't see private property being the issue. Speech has always been either based on private property (buy a printing press and start publishing) or public property (stand in the town square and start your speech and see if anyone has any interest). We're not designing anything to be private property; it's either private or it's public.
My right to free speech and my right to ignore anyone are equal rights to me. The only reason free speech is such an issue is because governments (and even corporations) so often try to suppress that right. That's been so extreme even hundreds of years ago that some people setting up a new government in the late 1700's decided to specifically address the issue that was a major problem. The right to ignore hasn't been a problem, but that doesn't make it any less of a right.
While the spammer at an ISP is spewing out millions, I get one or a few. Throttling them has no effect. I still get hit by hundreds of different IP addresses. If the realm of throttling is just one IP at a time, it won't work. If it covers all the IPs as a group, it has adverse affect on everyone.
And worse, this idea removes the incentive, because then the ISP won't be losing customers. There has to be "collateral damage" to make the boycott have an impact on an ISP. Maybe this is surprising to you, but lots of ISPs make the decisions about how they do things based on what will yield them the greatest revenues and profits. If they don't feel that their customers will leave because of some action, then they will have no reason not to do so if they feel it will bring in some revenue. And that action might be to accept a known spammer as a customer.
What other mechanism of causing customers to leave an ISP is there besides "collateral damage"? Spammers have a duty to stop spamming. Let me know when they stop so I can resume listening to free speech.
So if one browser gets better, and then because of the pressure another gets better, too, this is bad? Maybe we should remove some features from one of them to make another look better? It's sad that we would have to downgrade the capability of something before we are able to make a choice.
I rewrote my rc/boot scripts myself from scratch. I haven't had this problem for 3 years.
We need better standards among printers. Much of the problem is due to so many different kinds of printers, different drivers, different data formats. One single standard is needed and vendors must be force to comply.
I still haven't figured out how to do a number of things on my MS Windows 98 machine. For example, how do I tell Windows that my hardware clock runs as UTC and that it should still show me my local time.
This is more of a programming problem. Certain programmers think that they need to first erase the screen then rewrite it. Back before Linux, I wrote an editor for DOS, and I wrote my own screen window manager for it. The editor could simply open up window objects and update them much like curses, but simpler. When refresh was called, the screen was updated, but there was no flicker because it was never erased first. It simply updated everything, period. Parts that were changing content just changed. Parts that were not changing, didn't. And mine was so fast I could still do scrolling by full screen rewrites even on a 16 MHz machine.
Programming problem again. Teach programmers how to deal with the real world.
Why do you want access to my files? Leave me alone.
Like printers, this is a vendor problem. Find vendors who do a better job of not always changing the driver-to-hardware interface, and favor them over the vendors that keep screwing people over with the next board version. There is no reason every piece of hardware needs to have its own driver disk included, even for MS Windows (and this is a big cause of many system problems in Windows, too ... Bill Gates has said so).
What you are asking for is to show text as if it had newlines, when in fact it has none. Maybe you should be writing HTML instead of plain ASCII text. Don't mail it to me w/o newlines. But if you want to be able to reformat a range of text, maybe you should try emacs.
I just changed my resolution on the fly while entering this line of text by pressing the Ctrl-Alt-KeypadMinus combo. Then I pressed Ctrl-Alt-KeypadPlus to revert back.
If you don't boycott the bad ISPs ... the ones that keep signing up spammers who leave their previous ISP so they can have new address space not yet listed ... then spammers will constantly be showing up at new addresses, and listing them does little good. You have to boycott bad ISPs (the ones who harbor spammers) to make them either turn good, or move the good customers off to a better ISP. Of course there will be "collateral damage" ... but not to the ISPs that respond appropriately and cut off spammers. The non-spamming customers can complain to the ISP (this is part of the motive) or just move to another ISP.
Being listed in a blacklist is not the same as being indicted for a crime. It's a boycott. Different rules. Different standards. If you think that SPEWS, for example, has more collateral damage than you want your mail server to participate in, then don't use it.
Free speech is not about making sure everyone gets to say whatever they want to whoever they want however they want. It's about making sure government doesn't step in and limit it. Free speech cannot be used to justify stealing my printing press. Indeed, such a theft would deprive me of my right to free speech in such a case. Likewise, you have no right to steal my network server time for your own speech. I own it for mine (or that of my customers who pay me for its use). This is why it's a property issue. It's no tougher for me to deal with theft of my server time than dealing with theft of my printing press.
If you are communicating with me about matters we have a relationship based common interest in, then that is expected and fine. For example if you were to e-mail your reply to this post, that is reasonable to expect because by posting this (even if it wasn't in response to yours), there is an implied expectation of a response. Trying to sell me penis enlargers, golf balls, ink cartridges, and mortgages is not applicable communication, unless I asked (in some form) where I could find those things to buy.
If you really have written about better methods that don't generate false positives, then please show me, and please point out specific ways these methods work to drive incentives to ISPs to disconnect spammers and to not sign up at all known spammers. Spamming will not decline if ISPs have no counter incentive (to the revenues) to choose to not allow spammers in their network.
How else can you boycott the ISP w/o collateral damage? SPEWS does not list the ISP, and hence, no collateral damage, until the ISP has had plenty of time to cut off the spammer. In order to increase the level of pressure on the ISP, more of their address space has to be listed to "encourage" them to cut off the spammer. The usual first listing is the whole /24 the spammer is in (if they weren't doing it from the whole /24 in the first place). Maybe they will start listening once their own customers complain (and that's the proper place for the customers to complain to, their ISP). If they continue to ignore the problem, then eventually the whole ISP will be listed. If it's a multi-level ISP, their upstream starts to get listed, too.
The philosophy SPEWS appears to be using, and one I now agree with (previously I did not, but sometimes my opinions do change ... hey, I'm open minded), is that the spam problem will not go away by blocking only the spammers. ISPs have to play a part by not signing up known spammers, and cutting off spammers that got signed up because they were not known at first. Blocking spammers alone will be a never-ending battle because then there is no incentive for any ISP to turn them away and they just keep moving around to evade the blocking. To end spamming, the ISPs have to quit offering them services, or we have to quit accepting traffic from the set of ISPs that do harbor spammers.
It looks like collateral damage, but it's just another form of boycott. If I organize a boycott against my local newspaper, then the advertisers suffer because fewer people read their ads. And such boycotts are known to even extend to boycotting the advertisers if things get bad (and spam right has gotten very bad already). Is that fair to the advertisers? Of course not. But that's the nature of the activity; it is, among other things, trying to encourage the advertisers to cease advertising there. So in the same way, by boycotting a whole ISP address space, the idea is to encourage their customers to change to another ISP, until the ISP changes their ways.
Palestinians are not saying that Israel has no right to exist. What they are saying is that Israel has no right to taking Palestinian land (all 100% of it) as the basis for their existance.
What is needed to stop this is a moderating system which ranks the various traded products, as identified by their MD5 checksum signatures, according to some "measure of quality". By rank ordering, it cannot be used to entirely shutdown a trading network since everything would still be available. Products at 50 out of 100 would have received a ratio of good vs. bad moderations better than 50% of other products, and worse than the other 50% of products. It would not necessarily be a 50/50 good/bad moderation. Thus flooding of bad moderations across the board would have no effect, though it could be used to drive very specific classes of products down the list. But eventually, people would see the abuse and mod them back up. It would be sort of like moderation on slashdot, but everyone gets to play.
Now would it be possible to have selective moderation like slashdot has? Only a central authority could do that the way slashdot does. The big question would be judging who gets moderation points. As far as I know, on slashdot, it's almost entirely automated. With product trading, it would be harder to measure the quality by automation, so someone has to manually make the judgement calls and that brings some risks as well.
If individuals could be identified uniquely in some way, without the risk of exposing real identity, then meta moderation might work. One way to do that would be a slow rate of generating some kind of signed digital certificate that allows only so many to be generated at a time per network that receives it (and no personal identifying info included, and no records kept). Moderations and meta moderations would be signed by these anonymous certificates. You wouldn't know who moderated, but what you would know is that a group of moderations by the same certificate are probably from the same person and can be judged accordingly, good or bad. Excessive levels of moderation would also weaken your merit and derate your contributions.
I'm sure it's really megaBITS per second, otherwise the company would surely have promoted it as 24 megaBITS per second if the speed really was 3 megaBYTES per second (so they could use the larger number). You don't generally see 1.5 megabits per second promoted as 187.5 kilobytes per second. But this does show how reporters are still subject to making technical errors, which I suspect is due to their lack of knowledge of technical details. At least they got the case of Mbps correct.
Not all server environments can be just trivially made to be multithreaded and still work securely. Multithreading means more than just sharing the codebase. It means sharing the data and sharing the access permissions. It means a huge security exposure in servicing different users. So that is where the 1000 processes comes from (and its not a low estimate, either ... I just talked to one person the other day who had as many as 12000 processes running ... fortunately not JVMs ... during a peak of usage). If Java is to get beyond lame stuff like shopping carts, it needs to become wise to more isolated secure environments, and still do it efficiently.
1000 processes doing just-in-time compiling? Now tell me how that resultant machine code is shared securely between processes. If compilation to native machine code can be done, why not do so before run time, and then the stored machine code image can have its one copy in RAM shared by the many processes running. This wouldn't have been an issue with applets in a browser, but it certainly is in a server.
I like the Java language ... but this Java Virtual Machine is a hindrance. For things like downloaded applets that need to run in anyone's browser on any machine, a portable run-time is essential. But for building server-side applications, the JVM is the problem. So I'm looking for a compiler system with libraries that compiles Java to native machine code (with many platforms supported), and still does everything any Java program would expect ... except for things like being able to share the runtime code, both the executeable and the libraries, across thousands of process instances, and the raw speed native code gets. Oh, and GPL, LGPL, or BSD style licensing would be nice.
Nearly 200 different non-existant usernames in my various domains are being spammed, many on a continuing basis. These are usernames that have never existed, and never accepted delivery of mail, so they are definitely not confirmed opt-in's for anything. It just shows how far spammers will go, and how they never clean up their lists (as if that would help real people).
If you're only going to pay $50k for techs, then you're going to get the ones who can't spell. You get what you pay for. If you don't need to have techs writing reports, then the $50k techs will probably do just fine for you. If you want a tech that can also do things like write proper reports to management, and document the systems and programs they put together, then I suggest you start paying more so you get a better pool of applicants and can choose the ones that fit your needs. Don't whine about the porr quality of techs if all you're offering to hire them is a lousy salary level.
Unfortunately, Slashdot doesn't pick and choose its readers and posters, so of course you get all kinds here. But there are many out there who can spell, and who can write documentation (some who can just don't like to, though).
Having the government ban the spam content might well be in violation of the 1st amendment. Banning fraud (e.g. using someone else's identity, or a fictional one) isn't. That won't cut out spam, but it can at least narrow the scope of what originates in the USA (or another country if they have a similar law).
Of cource spam makes money. But not for everyone. There aren't enough consumers available for the practice of spamming to reach into the thousands of people doing what Scelson does. At that point the net just grinds to a halt from overload (despite being designed to resist military attacks during war).
You are so utterly ignorant of the law. Maybe that's why you are keeping yourself anonymous. Wanna cite the federal law you think exists to force businesses to do business with just anyone? You can't, because it doesn't exist.
As for anti-spammers, there is no interference whatsoever. They are simply boycotting. They have a right to boycott any business they choose. Oh, in case you hadn't noticed, they do have the right to refuse to do business with whomever they don't want to. And if they don't want to do business (trade packets) with an ISP that keeps spammers connected, that's their choice and their right. Now the ISP decides whether they want to do business with the spam side of the net, or the non-spam side. Seems more and more are choosing the non-spam side.
By making custom machines, local integrators do not have to lock themselves into those mass quantity deals that they need in order to make sure their huge volume stays consistent. This lets them pick and choose what's cheaper today, while Dell and the other big computer makers are stuck taking shipments from their long term contracts. Only if there comes to be a parts shortage will the situation reverse.
As for the service part, the one thing you left out is that when the local integrator does send a tech out for on-site service, it's more often the very same guy who therefore knows more about the problem history the customer might be having, and won't start over from the beginning every time like you'd get if Dell sends a different sub-contracted tech out each time.
Also, if the problem is due to some incompatible piece of hardware, the local integrator is much more free to use some different brand or model of part to replace it and get things working, while the sub-contracted tech from Dell has to stick with the Dell brand products (so the next tech will understand it).
While the unemployment rate of the US population in general is a mere 6%, among techies, unemployment plus underemployment is somewhere between 25% and 35%. Techies are not that big a portion of the whole population. With no disposable cash, of course they won't spend where it can be avoided. Of course there will be many who steal music even if well employed, but many others won't.
Impossible is coding 100000 different values into a 16 bit field. That is where "cheap" isn't one of the choices as you have to expand the field ... and that may mean recompiling and maybe even recoding everything else that expected to use just 16 bits.
Impossibility depends on how you state the problem. If the manager says "use a 16 bit field to code the 100000 values" then it is impossible. Tell them so and they might not believe you (in part because they may have come from a background where they are drilled with concepts like "nothing is impossible if you try hard enough", even though we know that is not true). Tell them 100000 requires at least 17 bits, and that its quicker to just use 32 bits, and then at least they haven't heard the word "impossible".
If you really do have to tell them it is impossible, maybe if you relate it to something they can understand, they can get the message. "That's like running a one minute mile". "A car can do that". "A car is a 32 bit field".