I have found that a couple simple procmail scripts will do the job. In general I assume; If it is sent to or from a known party, it gets routed. If not -- mailing list or bad routing header info -- it's probably spam.
This is definately a "good enough" style filter, and not intended to catch every theoretical variation of spam and good (but spam-like) messages. I find that it is rarely wrong (2 misrouted messages in 4 years).
Here is a summary of my filters, in order of execution;
Mark OK and move to a non-Spam folder:
If directly to me (not mailing list) or from a known person or list.
Move to Spam folder:
If a valid address is not in the To field.
If unknown mailing list with unknown senders.
If To: is invalid (blank, missing)
Move to Likely Spam folder:
If directly to known pseudo address that recieves spam.
The use of the word theory has already been commented on and I agree with that comment. If you disagree with that, there's no reason to read further.
Newtonian physics were useful and practical and are still valid for the study of larger objects. Even Einstein admited that his own theories were probably inaccurate and that even if his weren't he was refining Newton's work not negating it. Newton's observations still work in theory and in fact...just not on all scales. That said, science is mostly self-correcting and tentitive. The word theory simply emphasises this tentitive-ness and does not mean "I guess".
My main objection:Belief is a loaded word and does not mean "I observed clearly" but "I feel it is so". It was the "I feel" not the "I observe" that I was objecting to. "I feel" is necessary to make sense of the world in a practical way but has no place in a hard-nosed look at this subject. That requires observation. Breeding is only one of the drop-dead-obvious examples of evolution.
Evolution is plausable by itself because of the consistant, overwhelming, and non-contradictory evidence for it and has been hammered on for over a hundread years without a sucessful alternative. Many religous groups accept evolution as a fact...though this is a curiosity not a validation of the facts of observed evolution.
Creationism has no evidence for it beyond religous texts and the beliefs are shifting and different depending on the time, major religous group, or sect. Read my comments on goal-post moving in another thread for further comments on this.
If the above does not resolve any differences of opinion you and I have, I'll just have to leave it at "I respectfully disagree".
Package an unmodified Wine in with your Windows app.
Compile against unmodified Winelibs to port your Windows app.
Make changes to any part of Wine.
The only resonsibility anyone has under the LGPL is is to provide the modified LGPLed part of Wine to those who;
Ask for it.
Have recieved the binary (as a paid customer or if provided at no dollar cost).
Are willing to pay a nominal fee for the effort to provide the source (optional).
The only problem I see with this is if a company makes substantial changes to the LGPLed source, and they are unwilling/incapable to seperate the parts they want to keep for themselves into little propriatory modules, they would have an attitude problem.
Since patches to the LGPLed parts could be used as hooks to link in the propriatory modules, it does not seem like a dire problem for a half decient programmer. After all, they get the rest of Wine/Winelib for no dollar cost or effort.
Re:Important point from Joerg Mayer On Wine List
on
WINE May Change To LGPL
·
· Score: 4, Interesting
Transgaming already runs into this problem with thier own CVS of Wine (WineX).
If someone wants to build the latest WineX, they have to wait for Transgaming to release a binary; no CVS. People have asked for the Macrovision module to be broken out, but Transgaming have not been able to (yet?).
For those who haven't followed this, the complaint TG gives was that the copy restriction code needs to be patched in to various parts of WineX to get it to work. While I see this as a problem, it can't be a really big one.
The sticky issue is that providing a binary copy restriction module might cause problems with Macrovision Inc. -- the folks who provided this code (likely under a quite threatening NDA).
Can Transgaming make a seperate module...and will Macrovision like Transgaming's ideas well enough to allow it to be released? My bet is that Macrovision really don't want that part seperated from WineX. Right now, it's mixed in with a bunch of other code and is harder to understand. As a stand-alone module with hooks it would have a much higher chance that it could be easily thwarted on both Linux+Wine and Windows systems.
Personally, I *hate*, *lothe*, and *dispise* this type of thing. I have a few commercial non-game CDs that are useless largely because Macrovision's "Safedisc". Transgaming's version works...but only on a few CDs. Mostly, it doesn't. History keeps repeating...
All you're saying is the same argument that has been offered up for centuries. Each time we learn more and find out what fictions have been pushed as facts, the religious move the goalposts back and deny that a point has been scored.
In children, this attitude is cute and interesting. In philosophers, it's part of the trade. In adults making a reasoned argument, it's ingenuous and artificial.
...I believe in evolution. I also believe in creation...
Belief? I don't believe in evolution -- I wouldn't know how to do such a thing. Belief never comes into it.
The preponderance of the evidence leads me to an obvious conclusion -- changes in individual living things occur from generation to generation. Enough time and changes occur, and you have this thing called evolution. In some ancient businesses, it's just called breeding.
If that evidence wasn't there, I'd conclude differently...but not necessarily that a spirit or deity was the necessary other choice.
RAM is a mechanical device; even though it doesn't have joints and piviot points, the parts it does have do move and do wear out.
When's the last time you checked your RAM? I get about 1 bad module for every 2 machines. Defects usually show up on the initial test, though some don't show up for a few years.
Don't believe me? Try it yourself; Memtest86. I suggest running one full test (can take days) when you first build a machine, and when you run into odd problems that you can't figure out. The default tests are good, but I've had times where it did miss problems.
Yes, that explains why Unix is so secure. Thank goodness it was designed to use OpenSSH and shadow passwords so many years back. Can you imagine how hard it would be to "add" something like that later, like some feature?
Hmmm. I'll see your sarcasm and raise you a permissions restrictions -- per-user, per-daemon/process, and per file element of course. Multi-user means something in the Unix world. Along with Multi-user this thing called security did in fact get designed right in.
SSH (the protocol, not the project), PAM, and others clip right on to this well-tested structure...so, yes, all snippyness aside, that is one reason why Unix is so secure.
So, you see, those who tout Linux and decry Microsoft are really taking an ironic stance. They are helping MS (by hurting their competition) when they advocate Linux.
I've heard this before, and it does make sense on a few levels.
Here's another truth: While Linux matters in the short term, Linux doesn't matter in the long term. Linux by name might be around in 50 years, but if not a system spawned from a flavor of Unix will be around.
The same can't be said of Windows. With.Net, they're already trying to replace most of the existing APIs and much of the design -- and that's just the latest reworking. How much of the guts of DOS or Windows 1.x is the basis for Windows XP?
The reason for this is practical and unambiguous; any Unix-style OS and the programs that run on it are usually designed to be ported to different Unix flavors. It's an expectation. The user also doesn't have to relearn much on the software side even if the hardware is drastically different. At it's core, Unix may be obnoxious but it is also highly consistant even when there are substantal differences like SCO and OSX's file structure. DOS 1.24 (I had it) and Windows 2000 file structures and commands don't share anything except for the most minor of details.
To verify this design stability and portability takes half a second.
* For unix systems: basic commands, the design of X11, a portable kernel,... -- currently running on everything from PDAs to mainframes from multiple vendors.
* For Windows systems: Only one vendor. Even Microsoft's solitare runs on...well...x86 if you don't count the blip during the first few releases of NT. Even counting the different platforms, most apps (including MS Office) were poor ports that used x86 emulation layers.
Re:Wine will not make users switch to Windows
on
Wired Talks Wine
·
· Score: 2
Here's an analogy that you might find useful.
I had to repaint a few rooms the other day. At first glance, there were only a few spots and since I had the right paint I preped and painted just those areas. When the paint dried, those places were perfect.
Yet, dents and other spots that I didn't notice at all started to become obvious. So, I preped and painted those new areas. Guess what happened when the paint dried?:(
While I generally agree with what you say...it's way too dire. The time to move is when it's practical, not when every little defect has been addressed.
We should take a page from Microsoft's handbook and accept 80% perfection...as long as the remaining 20% isn't honestly critical for the tasks at hand.
In most cases, I've heard unknowledgeabe gripes about non-Windows environments even from those who are technically wizards under other forms of Unix. The same picking of nits about Mozilla still occurs -- even now that Mozilla rocks and has for a few months.
Reminds me of my nieces refusing to eat sushi because they think they know what it tastes like.
Well, even a Windows user would eventually have to upgrade apps and hardware too...if they upgraded Windows.
Quick fix: Get VMware. It works very well, but has different limitations compared to Wine; most non-3D accelerated programs work perfectly. Win4Lin is also good, and cheaper. It can run applications faster, but it only runs Windows programs, and a smaller number of those. Both require a copy of Windows, where Wine does not. The CD that came with the laptop (if any) may or may not work. Sites: vmware.com and www.netraverse.com.
For my effort, I've started to add applications -- even trivial things like screensavers -- to the Wine Application Database. I've included debugging output for apps -- if they work perfectly or fail drastically. I've included notes on how to get things working with Wine. I've borrowed CDs from friends and tried to install them under Wine.
Do the same, and you might be surprised. Most programs don't work under Wine, but a surprisingly growing list of programs do. Some do require hand-masaging, though most don't.
If more people report what they find -- good or bad -- Wine will get better for the general user.
People don't want "an office suite", they want OFFICE. They KNOW Office. Even if they don't necessarily LIKE Office, they know how to use it and don't want to learn something "like Office" unless it's EXACTLY "like Office".
That's what they say, it's not how they act if even the slightest deterant is put up against MS Office while StarOffice is the easy pre-installed choice.
What they complain about is change and any hassles they might have to put up with. My experience matches what Lumpy said in this thread;
OK, it's interesting as an experiment in Intel-specific compilers. As a baseline when testing out any new x86-compatable processors, it's probably a critical tool. But otherwise?
Seriously: Why even bother?
As for proclaiming GCC dead...please. Speed benchmarks and compilers are notorious PR pieces. I can't think of a better example of pure sensationalism. Disagree? Prove me wrong.
On a related note, there are 2 apps that I'd love to see on Linux, but only have Windows versions. The first is the Unabridged Oxford English Dictionary 2nd Edition on CD-ROM, and the other is that huge CD-ROM collection from National Geographic that spans from 1888-2000 (or there abouts). Quite a project, to be sure, and I've always wondered myself if I'd catch hell if I ever figured out the DB structures and released the code to the OSS community.
Hmmm. [punches in 'geographic' on freshmeat.net. Clicks on first link.] Do you mean something like this?
I can't say if this fits your need, but the beta release of Sun StarOffice 6.0 creates/opens/saves MS Word documents with revision marks. Seems to do so perfectly, though I honestly haven't stress tested it yet; I don't often edit documents with revision marks.
If I'm mistaken and some feature of revision marks doesn't work as expected let me know and I'll go bring it up in one of the OpenOffice lists.
As for the other issues, list them here or (better yet) tell the folks at OpenOffice.org. They are very open to comments on improving the editor as well as any other part of the suite.
Revision marks quick how-to:
New revision marks -- Edit...Changes...Record.
Export -- File...Save as... and choose one of these;
Microsoft Word 97/2000/XP
Microsoft Word 95
Microsoft Word 6
Import -- File...Open (choose document from list; default is show "All" files)
Since you are in journalism, I'll leave it up to you to get Sun to fork over a copy of StarOffice 6 beta. The last open beta closed at the end of 2001. The next release is expected to be 6.0 final (or close to it).
Re:Loki didn't work, but other things might:
on
Last Word on Loki
·
· Score: 2
Well, I agree with most of your points. There are major problems to handle using bood CDs for most modern games.
Many of these problems are not impossible to handle, though. Look at www.demolinux.org and the compression technique used by Knoppix www.knopper.net/knoppix (German -- translates OK).
The compression technique used by Knoppix, for example, allows about 2G of data to fit on a single 700MB CDR. The speed loss due to compression is somewhat made up for by the speed gain from more data being transfered from the CD at one time. This method is filesystem independent.
Contracts. Untill recently, AOL had to use IE if it wanted a spot on the Windows desktop. Since that contract has expired and was not renewed, AOL now has no motivation to stay with IE.
With the browser being a critical part of AOL's interface, the switch can't happen without care. Compuserve -- now a division of AOL/TW -- uses the Mozilla/Gecko/Netscape/... engine. Once they work out all the glitches there, switching the AOL user base over is a viable option. A few months ago, quite a few banks and financial sites didn't work with Netscape 6.x, so switching then would be a problem.
Doesn't mean that they will switch now, but I'd be surprised if IE is still used in a year.
Are you sure? The 1 billion I heard about was for 3 years. Still not shabby, but if it's really 1B/yr, I'd like to know about it! (Links if you've got em'!)
AOL, with Gateway, have used Linux in the past for embeded, AOL-branded, Internet devices.
Expect more of this in the future regaurdless of what (if anything) happens with AOL/TW and RedHat.
As anyone who has worked on commercial software knows, the release schedule drives the features list and the features list drives both coding and testing.
Security is one of those things that is required to come at the planning stage of any product -- not as an afterthought during the coding and test stages.
MS needs profits to buy new companies so they don't have to pay divedends. They need big profits so that the stockholders will be happy with the 'value' of MS as a whole.
Yet, the software side of thier business is a stagnent market -- huge and captive but not growing as it used to. Because of that they need to retain customers and get them to upgrade on a regular basis (subscriptions everyone?).
Then, we're back to the schedule and the features and security getting short shrift.
Slashdot Mirror
This is definately a "good enough" style filter, and not intended to catch every theoretical variation of spam and good (but spam-like) messages. I find that it is rarely wrong (2 misrouted messages in 4 years).
Here is a summary of my filters, in order of execution;
Mark OK and move to a non-Spam folder:
Move to Spam folder:
Move to Likely Spam folder:
The use of the word theory has already been commented on and I agree with that comment. If you disagree with that, there's no reason to read further.
Newtonian physics were useful and practical and are still valid for the study of larger objects. Even Einstein admited that his own theories were probably inaccurate and that even if his weren't he was refining Newton's work not negating it. Newton's observations still work in theory and in fact...just not on all scales. That said, science is mostly self-correcting and tentitive. The word theory simply emphasises this tentitive-ness and does not mean "I guess".
Evolution is plausable by itself because of the consistant, overwhelming, and non-contradictory evidence for it and has been hammered on for over a hundread years without a sucessful alternative. Many religous groups accept evolution as a fact...though this is a curiosity not a validation of the facts of observed evolution.
Creationism has no evidence for it beyond religous texts and the beliefs are shifting and different depending on the time, major religous group, or sect. Read my comments on goal-post moving in another thread for further comments on this.
If the above does not resolve any differences of opinion you and I have, I'll just have to leave it at "I respectfully disagree".
The only resonsibility anyone has under the LGPL is is to provide the modified LGPLed part of Wine to those who;
The only problem I see with this is if a company makes substantial changes to the LGPLed source, and they are unwilling/incapable to seperate the parts they want to keep for themselves into little propriatory modules, they would have an attitude problem.
Since patches to the LGPLed parts could be used as hooks to link in the propriatory modules, it does not seem like a dire problem for a half decient programmer. After all, they get the rest of Wine/Winelib for no dollar cost or effort.
If someone wants to build the latest WineX, they have to wait for Transgaming to release a binary; no CVS. People have asked for the Macrovision module to be broken out, but Transgaming have not been able to (yet?).
For those who haven't followed this, the complaint TG gives was that the copy restriction code needs to be patched in to various parts of WineX to get it to work. While I see this as a problem, it can't be a really big one.
The sticky issue is that providing a binary copy restriction module might cause problems with Macrovision Inc. -- the folks who provided this code (likely under a quite threatening NDA).
Can Transgaming make a seperate module...and will Macrovision like Transgaming's ideas well enough to allow it to be released? My bet is that Macrovision really don't want that part seperated from WineX. Right now, it's mixed in with a bunch of other code and is harder to understand. As a stand-alone module with hooks it would have a much higher chance that it could be easily thwarted on both Linux+Wine and Windows systems.
Personally, I *hate*, *lothe*, and *dispise* this type of thing. I have a few commercial non-game CDs that are useless largely because Macrovision's "Safedisc". Transgaming's version works...but only on a few CDs. Mostly, it doesn't. History keeps repeating...
In children, this attitude is cute and interesting. In philosophers, it's part of the trade. In adults making a reasoned argument, it's ingenuous and artificial.
Please snap out of it.
Belief? I don't believe in evolution -- I wouldn't know how to do such a thing. Belief never comes into it.
The preponderance of the evidence leads me to an obvious conclusion -- changes in individual living things occur from generation to generation. Enough time and changes occur, and you have this thing called evolution. In some ancient businesses, it's just called breeding.
If that evidence wasn't there, I'd conclude differently...but not necessarily that a spirit or deity was the necessary other choice.
Sid Meier's Alpha Centuri is worthy of it's acronym.
When's the last time you checked your RAM? I get about 1 bad module for every 2 machines. Defects usually show up on the initial test, though some don't show up for a few years.
Don't believe me? Try it yourself; Memtest86. I suggest running one full test (can take days) when you first build a machine, and when you run into odd problems that you can't figure out. The default tests are good, but I've had times where it did miss problems.
Hmmm. I'll see your sarcasm and raise you a permissions restrictions -- per-user, per-daemon/process, and per file element of course. Multi-user means something in the Unix world. Along with Multi-user this thing called security did in fact get designed right in.
SSH (the protocol, not the project), PAM, and others clip right on to this well-tested structure...so, yes, all snippyness aside, that is one reason why Unix is so secure.
I've heard this before, and it does make sense on a few levels.
Here's another truth: While Linux matters in the short term, Linux doesn't matter in the long term. Linux by name might be around in 50 years, but if not a system spawned from a flavor of Unix will be around.
The same can't be said of Windows. With .Net, they're already trying to replace most of the existing APIs and much of the design -- and that's just the latest reworking. How much of the guts of DOS or Windows 1.x is the basis for Windows XP?
The reason for this is practical and unambiguous; any Unix-style OS and the programs that run on it are usually designed to be ported to different Unix flavors. It's an expectation. The user also doesn't have to relearn much on the software side even if the hardware is drastically different. At it's core, Unix may be obnoxious but it is also highly consistant even when there are substantal differences like SCO and OSX's file structure. DOS 1.24 (I had it) and Windows 2000 file structures and commands don't share anything except for the most minor of details. To verify this design stability and portability takes half a second.
* For Windows systems: Only one vendor. Even Microsoft's solitare runs on...well...x86 if you don't count the blip during the first few releases of NT. Even counting the different platforms, most apps (including MS Office) were poor ports that used x86 emulation layers.
Let's see,
various programming tools,
games (from board to 3D hardware accelerated),
a couple of PIMs with handheld sync. support as well as a dozen email programs,
various Internet and network tools (from serious to trivial),
commercial software demos,
graphics packages,
personal finance,
...and a few office suites and assorted 'office' applications.
The only problem is that after a while, they start to run out of room on the box.
StarOffice 6 does support revision marks. Take a look at this comment.
I had to repaint a few rooms the other day. At first glance, there were only a few spots and since I had the right paint I preped and painted just those areas. When the paint dried, those places were perfect.
Yet, dents and other spots that I didn't notice at all started to become obvious. So, I preped and painted those new areas. Guess what happened when the paint dried? :(
While I generally agree with what you say...it's way too dire. The time to move is when it's practical, not when every little defect has been addressed.
We should take a page from Microsoft's handbook and accept 80% perfection...as long as the remaining 20% isn't honestly critical for the tasks at hand.
In most cases, I've heard unknowledgeabe gripes about non-Windows environments even from those who are technically wizards under other forms of Unix. The same picking of nits about Mozilla still occurs -- even now that Mozilla rocks and has for a few months.
Reminds me of my nieces refusing to eat sushi because they think they know what it tastes like.
Quick fix: Get VMware. It works very well, but has different limitations compared to Wine; most non-3D accelerated programs work perfectly. Win4Lin is also good, and cheaper. It can run applications faster, but it only runs Windows programs, and a smaller number of those. Both require a copy of Windows, where Wine does not. The CD that came with the laptop (if any) may or may not work. Sites: vmware.com and www.netraverse.com.
For my effort, I've started to add applications -- even trivial things like screensavers -- to the Wine Application Database. I've included debugging output for apps -- if they work perfectly or fail drastically. I've included notes on how to get things working with Wine. I've borrowed CDs from friends and tried to install them under Wine.
Do the same, and you might be surprised. Most programs don't work under Wine, but a surprisingly growing list of programs do. Some do require hand-masaging, though most don't.
If more people report what they find -- good or bad -- Wine will get better for the general user.
That's what they say, it's not how they act if even the slightest deterant is put up against MS Office while StarOffice is the easy pre-installed choice.
What they complain about is change and any hassles they might have to put up with. My experience matches what Lumpy said in this thread;
OK, it's interesting as an experiment in Intel-specific compilers. As a baseline when testing out any new x86-compatable processors, it's probably a critical tool. But otherwise?
Seriously: Why even bother?
As for proclaiming GCC dead...please. Speed benchmarks and compilers are notorious PR pieces. I can't think of a better example of pure sensationalism. Disagree? Prove me wrong.
Hmmm. [punches in 'geographic' on freshmeat.net. Clicks on first link.] Do you mean something like this?
Side note: Some notes on the StarOffice site about revision marks talk about "versioning".
If I'm mistaken and some feature of revision marks doesn't work as expected let me know and I'll go bring it up in one of the OpenOffice lists.
As for the other issues, list them here or (better yet) tell the folks at OpenOffice.org. They are very open to comments on improving the editor as well as any other part of the suite.
Revision marks quick how-to:
Export -- File...Save as... and choose one of these;
Microsoft Word 95
Microsoft Word 6
Import -- File...Open (choose document from list; default is show "All" files)
Since you are in journalism, I'll leave it up to you to get Sun to fork over a copy of StarOffice 6 beta. The last open beta closed at the end of 2001. The next release is expected to be 6.0 final (or close to it).
Many of these problems are not impossible to handle, though. Look at www.demolinux.org and the compression technique used by Knoppix www.knopper.net/knoppix (German -- translates OK).
The compression technique used by Knoppix, for example, allows about 2G of data to fit on a single 700MB CDR. The speed loss due to compression is somewhat made up for by the speed gain from more data being transfered from the CD at one time. This method is filesystem independent.
Details:
With the browser being a critical part of AOL's interface, the switch can't happen without care. Compuserve -- now a division of AOL/TW -- uses the Mozilla/Gecko/Netscape/... engine. Once they work out all the glitches there, switching the AOL user base over is a viable option. A few months ago, quite a few banks and financial sites didn't work with Netscape 6.x, so switching then would be a problem.
Doesn't mean that they will switch now, but I'd be surprised if IE is still used in a year.
Are you sure? The 1 billion I heard about was for 3 years. Still not shabby, but if it's really 1B/yr, I'd like to know about it! (Links if you've got em'!)
AOL, with Gateway, have used Linux in the past for embeded, AOL-branded, Internet devices. Expect more of this in the future regaurdless of what (if anything) happens with AOL/TW and RedHat.
Security is one of those things that is required to come at the planning stage of any product -- not as an afterthought during the coding and test stages.
MS needs profits to buy new companies so they don't have to pay divedends. They need big profits so that the stockholders will be happy with the 'value' of MS as a whole.
Yet, the software side of thier business is a stagnent market -- huge and captive but not growing as it used to. Because of that they need to retain customers and get them to upgrade on a regular basis (subscriptions everyone?).
Then, we're back to the schedule and the features and security getting short shrift.
Does anyone expect it to be any other way?
Windows: Focused on security since 2002. Really, we're serious this time. Stop laughing.