I've only got experience of this stuff with QMail, but for that MTA, there are patches to add X-RBL-Hit headers or similar, as well as the more common don't-accept-RBL-listed-connections type of thing.
I don't know if this is a specific qmail issue, but the way the server works is to accept the mail if it for a local domain, queue it and then figure out what to do with it in terms of the user to deliver to, since there may be aliases and so on involved, which it doesn't want to spend time resolving when accepting the mail. In the case where the spammer is just guessing usernames en masse with a domain name on the end, all the 'misses' become bounces. If spammers were at least halfway decent and mailed address known to exist, the load on mail servers from spam would probably be considerably reduced.
So the difference is that the ISPs mail server will then spend a large chunk of time generating bounce messages (which will typically also bounce back, as the return address of spam is often faked), rather than refusing to accept the single connection in the first place (a single connection can spawn hundreds or thousands of queued items - a large BCC list effectively).
Obviously, the time spent by the mail server clearing queues of bounce messages and double-bounces is time spent not delivering customers incoming and outgoing mail.
Personally, I agree with the RBL in it's DNS form, but not the BGP version - blocking routing to IPs and especially IP ranges completely is extreme and harms much more than the intended target.
Of course, if you like the idea but not the politics, you could reasonably easily set up a competing service - it's just DNS data. Maybe make it group-moderated in some way - slashdot for spam-prevention. *shudder*. I believe there are at least a couple of similar things for usenet spam (can't remember the names though).
Pantone Color is a proprietry colour-matching standard - you buy a license to allow you to use it in your app.
The theory goes that when you specify your colour as a Pantone number, your printer, or repro house can match that colour exactly because their ink manufacturer can sell them spot colours in those exact colours, rather than mixing up CMYK to get there. You can get a book of 'perfect' prints of these colours, and then use that to be sure that you're going to get what you want, because your screen is not going to give you WYSIWYG in colour terms.
Many people use particular Pantone colours in their letterhead/logo etc, and having a graphics app with support for it allows you to match their colouring for work you produce. It's more important for print than online work though.
but it didn't crash nearly as much as Photoshop did.
I can't remember the last time photoshop crashed on me - my only current problem seems to be a driver bug that both Painter and Photoshop tickle in my GeForce... big blits in certain situations kill the machine completely. On my other system, PS is rocksolid though - has been since v4 or so.
Aside from the cost to me in recieving it, the time involved in setting up some sort of filter for it, or in deleting it?
Aside from that, it's the apparent stupidity of the senders - being sent a ton of mail that is either barely english, or doesn't even give you a sensible way to actually buy the product that it purports to be offering, or is just a plain scam.
Oh, and people who don't filter their address-list for people in other countries who just can't take advantage of that one-time only offer (SAVE $$$$) anyway - that's just wasting everybody's time and money. Canadian life-assurance companies, or US mail order by 800 numbers are particularly pointless.
and spam for spam-agencies. I have had about a dozen for the same people in the last three days.
So it's more the mentality of the senders that pisses me off, and what that implies about the way they regard their 'customers'.
From my memory of Usenet, a bear is a bearded man (I think there was a more specific part to it too - big? uniformed?) - they come in all creeds and colours.
Cashing in on the literacy on some of their customers, I guess could also be "bear-naked black chicks".
You`re free to cross elsewhere if you`re that sad!
That would depend on your location - I understand that Jaywalking is taken a lot more seriously in some countries relative to others (Germany is one such, IIRC).
Besides, having tried walk to/from places in a few US towns (Tucson, DC suburbs) it's at best a challenge to cross elsewhere, if not downright dangerous/physically impossible.
I hate being a pedestrian on holiday. More, I hate being required to drive in order to get from one side of town to the other.
A very good majority of the games which ran on Win95 -- any version -- will still run on any of those platforms you mention... like Windows NT, Win98, WinME, etc.
This is certainly not my experience! Many games "require" DirectX 5 (*) which was just not available for NT 4. Better yet, a fair number of those games don't work on Win2k because they check for the OS not for the DLLs ("X doesn't work on NT because it required DX5"), even though Win2k does have a full DirectX 7 install, including Direct3d finally. Even a bunch of OpenGL games use DX for either UI or Sound/Control stuff, so they don't work either.
With all of that said, I'd still rather develop my (crappy little) games on Windows than Linux - the big mess of linux 'nearly' things is too annoying: it nearly supports hardware 3d, it nearly supports some soundcards, it nearly supports USB game controllers... maybe I'm just unlucky with my choices of hardware.
(*)the reason seems to be largely because it was the first version to have DirectInput, so you can deal with controllers in a standard way.
My QM is a bit rusty - it's what finally made me realise I shouldn't be doing a physics degree in 1992ish - but isn't this "really hard" to use a technical term?
I thought the difficulty of measuring spin on electrons without losing the spin in the process was what made quantum channels so great for secure transmissions...
I thought the idea was that we'd put up sites to hawk our wares, with product reviews by trusted individuals, register our pages with search engines and let people find us when they want us.
Yeah, because that way works so well. Case in point: this afternoon, I spent 15 minutes or so trying to find a company that I knew existed, and that has at least two websites(one in the US, and one in the UK) using search engines - no dice. It's a nice Schematic Capture, PCB design and circuit emulation package that can also emulate some microcontrollers - you can build an all-virtual PIC board, for instance. The product has a funny name I don't remember. I want to give these people money (the product is cheap, and has a nice modular purchasing method), but I can't.
This may partly be the fault of someone having 'creative' product names, but the current search engines don't allow people to find products when they want to.
It's not often I click on banner ads, but that is generally because they are largely for either borderline-scams (lotteries, casinos etc) or crap (internet connection 3x faster junkware). On the occasion that I am presented with an advert for an interesting product (thinkgeek manage this reasonably often on/.) - I go and have a look. It's only the extremes that are flat-out annoying: dozens of popups, and more ad-space than content-space.
On the other hand, for many folks, the site is the product - what do they do in your no-ad utopia? Are you also in the camp where payment for content on the web is also inherently evil?
Anyone who calls for "open source or anything not from Microsoft" has no ideals worth mentioning.
Since when did using a computer become something requiring ideals? If you want to be an free-software purist, then good for you. I just want my PC to work right and do what I need. Sometimes open-source or free software does that for me, sometimes it doesn't.
As I understand it, the hard-FSF line is that if I can't do it with free software, and I can't write the software to do it myself, then I should sit on my hands and find something else to do while waiting for the FSF crusaders replacing every application with a free one to get around to mine, and advocate free software to commercial software houses while I'm waiting. This is not particularly practical for solving my original problem.
Claiming that as a result that I am selling out in some way is as irrational as the 'anything but BG' line - it's a tool, not a cause.
I actually agree with a lot of the FSF's arguments, but a little pragmatism is needed.
Because in LaTeX you normally do define the document in terms of it's structure (/section{Introduction}) and then redefine what/section{Introduction} means in terms of dots on paper, I would expect it to be easier if anything.
CSS is available enough now that HTML 'designers' just might start doing that more, using <h1 class="SectionHeading"> rather than <font size=6 color="Purple" face="Arial">Introduction</font><br><br><br>.
That said, while the default style of LaTeX is fairly attractive, it's pretty tricky to get out of - which annoys those with intact vision.
(season comments to taste - I haven't used LaTeX since about 1992, when Latex2e was just arriving, and I gave up Physics - I believe 2e sorted out a lot of problems with controlling the look, including sensible PS font support)
So, having trashed NT administration for the mess it is (no serious disagreement there, although it's mostly worked OK for me in the past, I can't see it scaling well), how do you "easily" admin 6000 unix boxes then? ssh is not an answer...
The way most people seem to suggest is to reduce the amount of unix there is by making the workstations as thin as possible (NFS mount everything, little/no local storage, little/no local apps) - you might as well use Xterminals (if you can find someone who still sells them) or Citrix (for which there are plenty of dedicated clients)...
When the printing press was invented, did they figure out how to prevent people from copying books by watermarking?
When the printing press was invented, there was no way of copying the image of the printed book. You either got your own printing press and re-set the whole thing, or hired a bunch of monks. In a tortured music analogy, they are the equivalent of learning to play stairway to heaven yourself on your guitar, or telling someone to play you that song that goes da-da-da-daaaa-dada-dooo.
If the slashdot poll was anything to go by, a sizable chunk (about a third) of/. folks aren't from the US. I'm not. However, I don't think that the Canadian elections are any more News For Nerds that the US ones. That's what cnn.com (or whoever) do. Then again, I couldn't give a stuff about what Anime will appear on Cartoon Network US either. That's what the user prefs page is for.
wasn't the dragon book set in TROFF or something equally vile?
If you mean Principles of Compiler Design, then it is/was published by Addison-Wesley, not O'Reilly, even with the animal name:-)
IIRC, W Richard Stevens wrote all his books in troff too - it was originally designed as a writer's tool, so it can't be that bad - I always used LaTeX instead simply because I was writing physics papers and LaTeX does math and structured documents so well.
Slashdot Mirror
I've only got experience of this stuff with QMail, but for that MTA, there are patches to add X-RBL-Hit headers or similar, as well as the more common don't-accept-RBL-listed-connections type of thing.
I don't know if this is a specific qmail issue, but the way the server works is to accept the mail if it for a local domain, queue it and then figure out what to do with it in terms of the user to deliver to, since there may be aliases and so on involved, which it doesn't want to spend time resolving when accepting the mail. In the case where the spammer is just guessing usernames en masse with a domain name on the end, all the 'misses' become bounces. If spammers were at least halfway decent and mailed address known to exist, the load on mail servers from spam would probably be considerably reduced.
So the difference is that the ISPs mail server will then spend a large chunk of time generating bounce messages (which will typically also bounce back, as the return address of spam is often faked), rather than refusing to accept the single connection in the first place (a single connection can spawn hundreds or thousands of queued items - a large BCC list effectively).
Obviously, the time spent by the mail server clearing queues of bounce messages and double-bounces is time spent not delivering customers incoming and outgoing mail.
Personally, I agree with the RBL in it's DNS form, but not the BGP version - blocking routing to IPs and especially IP ranges completely is extreme and harms much more than the intended target.
Of course, if you like the idea but not the politics, you could reasonably easily set up a competing service - it's just DNS data. Maybe make it group-moderated in some way - slashdot for spam-prevention. *shudder*. I believe there are at least a couple of similar things for usenet spam (can't remember the names though).
Pantone Color is a proprietry colour-matching standard - you buy a license to allow you to use it in your app.
The theory goes that when you specify your colour as a Pantone number, your printer, or repro house can match that colour exactly because their ink manufacturer can sell them spot colours in those exact colours, rather than mixing up CMYK to get there. You can get a book of 'perfect' prints of these colours, and then use that to be sure that you're going to get what you want, because your screen is not going to give you WYSIWYG in colour terms.
Many people use particular Pantone colours in their letterhead/logo etc, and having a graphics app with support for it allows you to match their colouring for work you produce. It's more important for print than online work though.
but it didn't crash nearly as much as Photoshop did.
I can't remember the last time photoshop crashed on me - my only current problem seems to be a driver bug that both Painter and Photoshop tickle in my GeForce... big blits in certain situations kill the machine completely. On my other system, PS is rocksolid though - has been since v4 or so.
Using an OTP system with a software implementation will get rid of your giant keyfob at least...
I would take Tucson over Laurel, MD or Delphi, MD any day of the week.
I like college towns. I liked Arizona in general too.
but all responsible gun nuts^H^H^H^H users^H^H^H^H^H^H owners have a locked gun cabinet - like the blasting caps.
Aside from the cost to me in recieving it, the time involved in setting up some sort of filter for it, or in deleting it?
Aside from that, it's the apparent stupidity of the senders - being sent a ton of mail that is either barely english, or doesn't even give you a sensible way to actually buy the product that it purports to be offering, or is just a plain scam.
Oh, and people who don't filter their address-list for people in other countries who just can't take advantage of that one-time only offer (SAVE $$$$) anyway - that's just wasting everybody's time and money. Canadian life-assurance companies, or US mail order by 800 numbers are particularly pointless.
and spam for spam-agencies. I have had about a dozen for the same people in the last three days.
So it's more the mentality of the senders that pisses me off, and what that implies about the way they regard their 'customers'.
From my memory of Usenet, a bear is a bearded man (I think there was a more specific part to it too - big? uniformed?) - they come in all creeds and colours.
Cashing in on the literacy on some of their customers, I guess could also be "bear-naked black chicks".
You`re free to cross elsewhere if you`re that sad!
That would depend on your location - I understand that Jaywalking is taken a lot more seriously in some countries relative to others (Germany is one such, IIRC).
Besides, having tried walk to/from places in a few US towns (Tucson, DC suburbs) it's at best a challenge to cross elsewhere, if not downright dangerous/physically impossible.
I hate being a pedestrian on holiday. More, I hate being required to drive in order to get from one side of town to the other.
(end of off-topic rant)
uh, 8+ years ago was before all this web-crap popped up already. Proactive nostalgia?
oops - you're right - been a while since I saw it.
Whistler in "Sneakers"
Win2k (with its 40 million lines of code -- 500% bigger than Linux -- and still called "micro").
40 million in the kernel, or the OS? The OS includes an awful lot of extra goodies beyond the kernel... IE, Explorer, Notepad, that crappy pinball...
The LCARS e-theme would help with interoperability at the UI level at least.
A very good majority of the games which ran on Win95 -- any version -- will still run on any of those platforms you mention... like Windows NT, Win98, WinME, etc.
This is certainly not my experience! Many games "require" DirectX 5 (*) which was just not available for NT 4. Better yet, a fair number of those games don't work on Win2k because they check for the OS not for the DLLs ("X doesn't work on NT because it required DX5"), even though Win2k does have a full DirectX 7 install, including Direct3d finally. Even a bunch of OpenGL games use DX for either UI or Sound/Control stuff, so they don't work either.
With all of that said, I'd still rather develop my (crappy little) games on Windows than Linux - the big mess of linux 'nearly' things is too annoying: it nearly supports hardware 3d, it nearly supports some soundcards, it nearly supports USB game controllers... maybe I'm just unlucky with my choices of hardware.
(*)the reason seems to be largely because it was the first version to have DirectInput, so you can deal with controllers in a standard way.
They have - the bugs are in copyrighted code, therefore the bugs are copyrighted.
The existence of them isn't though... reproduce in this context isn't "make a copy" but "produce the same symptoms".
My QM is a bit rusty - it's what finally made me realise I shouldn't be doing a physics degree in 1992ish - but isn't this "really hard" to use a technical term?
I thought the difficulty of measuring spin on electrons without losing the spin in the process was what made quantum channels so great for secure transmissions...
I thought the idea was that we'd put up sites to hawk our wares, with product reviews by trusted individuals, register our pages with search engines and let people find us when they want us.
/.) - I go and have a look. It's only the extremes that are flat-out annoying: dozens of popups, and more ad-space than content-space.
Yeah, because that way works so well. Case in point: this afternoon, I spent 15 minutes or so trying to find a company that I knew existed, and that has at least two websites(one in the US, and one in the UK) using search engines - no dice. It's a nice Schematic Capture, PCB design and circuit emulation package that can also emulate some microcontrollers - you can build an all-virtual PIC board, for instance. The product has a funny name I don't remember. I want to give these people money (the product is cheap, and has a nice modular purchasing method), but I can't.
This may partly be the fault of someone having 'creative' product names, but the current search engines don't allow people to find products when they want to.
It's not often I click on banner ads, but that is generally because they are largely for either borderline-scams (lotteries, casinos etc) or crap (internet connection 3x faster junkware). On the occasion that I am presented with an advert for an interesting product (thinkgeek manage this reasonably often on
On the other hand, for many folks, the site is the product - what do they do in your no-ad utopia? Are you also in the camp where payment for content on the web is also inherently evil?
Anyone who calls for "open source or anything not from Microsoft" has no ideals worth mentioning.
Since when did using a computer become something requiring ideals? If you want to be an free-software purist, then good for you. I just want my PC to work right and do what I need. Sometimes open-source or free software does that for me, sometimes it doesn't.
As I understand it, the hard-FSF line is that if I can't do it with free software, and I can't write the software to do it myself, then I should sit on my hands and find something else to do while waiting for the FSF crusaders replacing every application with a free one to get around to mine, and advocate free software to commercial software houses while I'm waiting. This is not particularly practical for solving my original problem.
Claiming that as a result that I am selling out in some way is as irrational as the 'anything but BG' line - it's a tool, not a cause.
I actually agree with a lot of the FSF's arguments, but a little pragmatism is needed.
Because in LaTeX you normally do define the document in terms of it's structure (/section{Introduction}) and then redefine what /section{Introduction} means in terms of dots on paper, I would expect it to be easier if anything.
;<br><br>.
CSS is available enough now that HTML 'designers' just might start doing that more, using <h1 class="SectionHeading"> rather than <font size=6 color="Purple" face="Arial">Introduction</font><br>
That said, while the default style of LaTeX is fairly attractive, it's pretty tricky to get out of - which annoys those with intact vision.
(season comments to taste - I haven't used LaTeX since about 1992, when Latex2e was just arriving, and I gave up Physics - I believe 2e sorted out a lot of problems with controlling the look, including sensible PS font support)
So, having trashed NT administration for the mess it is (no serious disagreement there, although it's mostly worked OK for me in the past, I can't see it scaling well), how do you "easily" admin 6000 unix boxes then? ssh is not an answer...
The way most people seem to suggest is to reduce the amount of unix there is by making the workstations as thin as possible (NFS mount everything, little/no local storage, little/no local apps) - you might as well use Xterminals (if you can find someone who still sells them) or Citrix (for which there are plenty of dedicated clients)...
What about 6000 "real" unix systems?
When the printing press was invented, did they figure out how to prevent people from copying books by watermarking?
:-) )
When the printing press was invented, there was no way of copying the image of the printed book. You either got your own printing press and re-set the whole thing, or hired a bunch of monks. In a tortured music analogy, they are the equivalent of learning to play stairway to heaven yourself on your guitar, or telling someone to play you that song that goes da-da-da-daaaa-dada-dooo.
(I agree with the rest of your post though
If the slashdot poll was anything to go by, a sizable chunk (about a third) of /. folks aren't from the US. I'm not. However, I don't think that the Canadian elections are any more News For Nerds that the US ones. That's what cnn.com (or whoever) do. Then again, I couldn't give a stuff about what Anime will appear on Cartoon Network US either. That's what the user prefs page is for.
Have a look at Visicheck and see what your site looks like to those with (among other things) red/green colour deficit.
wasn't the dragon book set in TROFF or something equally vile?
:-)
If you mean Principles of Compiler Design, then it is/was published by Addison-Wesley, not O'Reilly, even with the animal name
IIRC, W Richard Stevens wrote all his books in troff too - it was originally designed as a writer's tool, so it can't be that bad - I always used LaTeX instead simply because I was writing physics papers and LaTeX does math and structured documents so well.