200 users is NOTHING. Until you are processing hundreds-of-thousands of messages per hour, you don't know how difficult it is to stop spam.
You want to know the really scary thing?
Both you and grasshoppa are right... for the subsets of the spam problem that you have to address.
My credentials? I run the mailfilters at a university with +50k student addresses, and around 3000 staff addresses. We typically reject a couple of hundred thousand messages daily. So, while our situation would probably turn grasshoppa's hair grey, it's a drop in the bucket compared to yours.
Spam is reasonably surmountable provided you have enough resources to throw at it. However, sometimes that doesn't scale too well
For example, I have two external relays load-balanced and ticking away running a whole lot of RBL checks, SpamAssassin checks and even using the Sane Security add-on signatures for CLAM while they do the regular virus scans. They presently have a whole lot of headroom, and if that ever diminished I could probably get a few thousand dollars for another box to share the load without too many questions given how well our solution works for us and how little it is costing otherwise. I imagine grasshoppa would have similar success with his or her solution. If you're talking hundreds of thosands of messages per hour, scaling my approach out to process that kind of message volume on a corporate budget dictated by the returns your bosses require could be interesting to say the least.
I'd add another item to the list, which some might see as a part of "No Magic Bullet", but I see it as sufficiently important to warrant its own entry - "There is no one-size-fits-all solution" . What works well-enough for you or me or grasshoppa will probably work well enough for similiarly sized organisations with similar resources and similar threats and similar values of "work well enough" - change any of that, though, and it's a whole new problem. The spammers are probably laughing their arses off over the kind of kind of religious wars people have over whose spamfiltering solution is best (it's mine, by the way, but then I also use vi to edit my conf files so of course it would be). And diversity of solutions is probably a good thing, as it means the spammers need to work harder to get past a whole lot of different approaches to get their garbage out.
It'll still tie the printout to your printer when The Man executes a warrant to examine all your printers. So you'd better be damn sure you haven't done anything that'll lead investigators to suspect you, and haven't sent any correspondence to government departments - donning my tinfoil hat here, what is there to stop governments routinely examining correspondence for these markings and linking senders to particular printers?
Has anyone reverse-engineered the watermarks sufficiently to enable someone to add bogus watermarks to printouts from non-watermarking printers? That could be fun, once RFID is sufficiently commonplace - walk past a stack of printers in a store, capture their serial numbers and give them to your anarc^bastard friends on the other side of the country who then procede to create bogus watermarks that get used to send threatening letters over the next year or two and get law enforcement hassling innocent people whose only crime was to buy a printer and send in the warranty card.
The idea is that this will not be visible to the naked eye - you should be cheering this announcment as a way to get rid of the barcodes that you hate but still keep the information.
Um... I thought that was exactly what I was doing, while also pointing out a possible problem with certain kinds of image. Things might get interesting if you're embedding patterns of yellow in an image that consists of a uniform white, or - for that matter - any other uniform or near-uniform colour. I suspect that under some circumstances it WOULD introduce visible artifacts - it would need to shift the yellow balance in sufficiently large blocks for crappy cameraphones to be able to pick it up, so if you're adding that to a solid white or some other solid or near-solid colour it may be visible.
(and who the hell modded me OT? Did they actually RTFA? And do they still have enough modpoints to come back and mod this "Flamebait"?)
Am I the only one who's annoyed by bar codes on CD covers and books?
Of course, this probably wouldn't fare too well on a re-issue of the White Album...
You're rant is lengthy but I think the gist is this. The free software community should accept binary drivers because you don't see a problem with "free as in no freedom to modify"?
In some cases, no - I don't. Do you have the right to modify your semi-automatic rifle to full-auto? If not, why the hell not? It's yours, why shouldn't you be able to do whatever you want with it? Why can't you remove the catalytic convertor and fiddle with the engine control systems for your car to increase its mileage dramatically - it's yours, and it's not your problem if certain kinds of emissions go through the roof. There's no problem if your softmodem scrambles the telephone exchange, or your modified wifi drivers screw things up for your neighbours, is there? Sometimes, there are very good public policy - and even just common-sense and common-courtesy - reasons to curtail what we can do. I can't piss on my neighbours at random, so why should my hardware be permitted to?
Of course, we may have some areas we agree on. If someone wanted to modify their display drivers to allow them to play HDCP content through their own capable but ageing projector or a DVI-only flat-panel, I see no reason why they shouldn't be able to do so. If someone wanted to modify their wifi drivers in a way that gained them some advantage but did so to the detriment of their neighbours, they ought not consider it in the first place - but because some people will, I have no problem with things that make it difficult for them to do so. Some people simply don't give a shit about what effect their actions will have on others, and so we need laws and rules and maybe the occasional piece of mob justice or political activism.
Free software isn't about price, why aren't you using Windows or OSX?
For some things, I am. For me, it's a matter of using a suitable tool for the task at hand. Don't I have the freedom to choose the tool that fits my circumstances best, even if I have to pay for it? As someone else got modded troll for saying, it sometimes seems like the FSF wants everyone to have the freedom to do things the way the FSF wants them to be done. And if I dislike something that a software vendor is doing, I have the freedom to look at alternatives and vote with my dollars - and even to write something that suits my needs better than some commercial offering and release it under whatever licence conditions I like.
I don't understand why we can't use proprietary drivers if they exist. I mean support from the hardware manufacturers are what Linux lacks and needs and what many wants, at least bitch about. Let proprietary and open source live together and take advantage of each others existence since proprietary drivers means that developers have one thing less to do and might use their time onanother project.
Additionally, for some devices with binary-only drivers, there are other matters that prevent the manufacturers opening up the drivers fully. The one that everyone always brings up first is that of the hardware manufacturers not always owning all the IP that goes into the drivers, and thus not being able to release all the specs. That's perhaps a very real and legitimate concern, and ought to be respected by people who get all pissy when some piece of free code gets misappropriated. The one I'd like to focus on though, even if it's slightly off-topic given that the discussion is video drivers, is that of open-source drivers not necessarily being compliant with regulations conceringing the operation of a particular piece of hardware. I've have EXTREMELY good results from an LT Winmodem (from the junkbox) I've dropped into a smoothwall box (also from the junkbox) that I built to save a relative from pr0ndiallers, but I would not like to see completely open-source drivers for such devices. "Why not?" you ask, quite legitimately. Well, who is most qualified to write drivers that affect things like how strong a signal and what encoding to pump down the phone line - some random hacker, or someone who knows PSTN technology? And who is liable if a home-built driver causes the modem to do nasty things to the phone line? At least with a vendor-supplied driver you can point the finger back to them, but here in Oz with a homebrew driver you could be facing a $12K fine if your softmodem does something funky to the exchange. With some wireless chipsets where a lot of stuff is done in software the same arguments could apply - it would theoretically be possible for people to write their own drivers that push the gear out-of-spec, which may be fine if you're out in the middle of the desert but which is definitely not desirable if you have to share the spectrum with your neighbours.
Given my 'druthers, I'druther have hardware that just does what it's asked to do with minimal CPU effort required - "Here are these bits, you know what to do". But then, I do recognise that software-based control of those devices means that they can be easier to update - for example, installing new drivers rather than flashing and hoping that you don't brick the device. I also recognise that CPUs are fast enough these days for the additional load from these kinds of devices to be neglible compared to that from the eye-candy most people have on their desktops these days, so the old system performance arguments usually don't hold water any more. Finally, from a freedom perspective, what's the difference between a black-box that you plug into your computer and a black box that you run on your computer? You don't hear people complaining that they're somehow oppressed because they can't get the source code for their external throw-bits-at-it-down-the-serial-line modem, yet somehow a software modem or wireless nic with binary-only drivers is the spawn of satan?
I don't care - mod me down if you feel compelled to do so. Honestly, though, I can't find a reason to get my knickers in a twist about binary-only drivers and their supposed deleterious effects on free software. If a manufacturer is prepared to put together good, stable and functional binary drivers, then good for them - they'll at least be considered next time I need a piece of hardware. If it gives me the freedom to choose from a wider range of products at a wider range of prices, I'm all for it. If you're concerned with the gear getting orphaned, think about the last time you used a piece of older gear - and what you used it in. Sure, I can p
Okay, that's a good prediction, but here's another in support of yours.
In two years, Microsoft will buy CodeWeavers. Each MS application will ship with the then-current version of CrossOver, for Linux and MacOS. Some time after that, Microsoft will do a premium distro that includes five years worth of service packs and CrossOver updates and their own desktop, so there's still scope for MS to do OEM and retail deals and continue to own the OEM space. Those who want to run MS apps under their favourite distro or even a roll-your-own installation can do so with a compatibility layer sanctioned and produced by Microsoft, said layer being paid for by the increased market share Microsoft gain and MS's reduced costs from being able to leverage outside community development. Okay, I kind of gagged on that last bit too, but I can see MS putting effort into fixing various kernel and system software issues and doing performance tweaking so their software works better, and under the GPL they'd need to give that stuff back. Their desktop they could licence commercially, of course, but it'd be in their interests to make sure anything that made their software work better on Linux got distributed as widely as possible. If MS got behind allowing their apps to run on Linux, that could only be a good thing for Linux in general.
It's a good idea - SMTP connections coming from a windows box that's not running NT/2k/2003 can't possibly be legit.
Unfortunately, that's not necessarily true.
There are a number of smaller businesses out there with something like Mercury32 or MDaemon running on Win95/98/ME, with halfway-decent firewalls that keep the bad guys from attacking directly or no attackable services running, and no web browsing from those boxes to expose them to the various web-based exploits that affect the out-of-date browsers on their machines. These set-ups were probably installed years ago by various consultants, and have been left alone because they continue to do what they're meant to. It's possible for something to have a Win98/98/ME fingerprint, legitimately send mail directly to your servers and be no threat to your or your users.
Harrison Ford took the Indiana Jones role over the Han Solo one because it's going to be a much better movie.
The rumour that it takes place in the sixties is true, and fits in nicely with the Mr Ford's present age.
What hasn't been widely revealed is that Sean Connory *will* be in the movie, although the role will probably surprise many people.
Since Satan owns the pink slip for the soul of pretty much everyone who has ever worked in motion pictures, he can shuffle the deck however he sees fit... and some interesting studio mergers mean that Sean Connory will play an elderly James Bond who fell through a temporal rift as the result of Xindi interference with Earth history - the theory being that if they could get all the kids hooked on beer and acid and dope then warp drive would never be invented. Little did they realise that Optimus Prime would ride in on My Little Pony and save the day by assassinating Kennedy and illegitimately fathering Rosie ODonnell with, you guessed it, Rosie ODonnell - who fell through the same temporal rift James Bond fell through. Pygmies re-discover left-over gou'auld technology that permits them to build hypersonic blow-dart weapons, which are capable of destroying ICBMs and thereby save the USA from the tyranny of total destruction when they decide to make the Ukraine glow in the dark...which happens two-thirds of the way through the movie, because the Ark of the Covenant (which was stolen from Area 51 by the Xindi) has been given to the Russians, who are using it to try to re-animate a cut-n-shunt SuperPolitician they've made from the cryogenically preserved remains of Adolph Hitler, Josef Stalin and Walt Disney - but exposure to nuclear fallout causes this re-animated monstrosity to sprout wings and fly to Tokyo, where as Mothra it does battle with Godzilla until Indiana Jones...
Sorry, I've given too much away already. You'll just have to buy a ticket like everybody else.
My wife used to work for a bank, and in the early 90's they were ditching the beta recorders they had used for internal corporate propaganda and standardised on VHS. One recorder, a whole stack of very-little-used cassettes, and a few longer tapes purchased brand new were used until about five years ago as a secondary unit (watch one channel, record two others - believe it or not, that was occasionally worthwhile) when the recorder died and it wasn't worth fixing.
Its picture quality was far superior to the VHS recorder we had at the time, but I have to wonder if that was due to the fact that it clearly wasn't as "built to a price" as the VHS units we could afford - although that could have been in part due to the fact that the bank had used high-grade tapes and the only things I could buy new for it were high-grade, and I wasn't about to spend $10 on a VHS tape for one-off viewing when I could get three lesser-grade ones for the same price. I wonder how much of the perceived quality difference between VHS and Beta has been shaped by the quality of the equipment and media used in the comparisons? Lounge rooms are hardly a laboratory environment, and it's reasonable to expect a mid-eighties Sony beta recorder with good tapes to outperform a 1990 Korean VHS cheapie with budget tapes.
I use RBLs. I like RBLs - most of our rejections are due to them, with SpamAssassin and the Sane Security signatures for CLAM responsible for most of the rest. When you reject a quarter million messages per day, and have no prospect of getting money for either the extra grunt or extra bandwidth required to analyse everything, it's a practical first line of defence.
That said, I'd been considering removing ORDB from our checks for some time. On days when NJABL and SpamHaus were picking up 30-50k messages each, ORDB would pick up between one and five messages. So, although it's sad to see their passing, for me at least they weren't that important a part of my spam-minimisation strategy.
(Interesting side note... while in an internet cafe here I happened to glance at the screen of the person working next to me. Guess what he was working on? A 419 email... So interesting to see it from this side!)
I am not in a position to fund your project, but can I send a few dollars for the removal of one or more of this person's fingers?
On a serious note, though, it would be very interesting to see some information on the kind of things that drive people into that kind of activity. I can't honestly say what I would or would not do if faced with extreme poverty and a low-risk data-entry position like that.
Here's an easy solution to being tailgated that ALWAYS works for me
You either drive a bigger piece of shit than I do, or you enjoy doing battle with insurance companies.
About twelve years ago, I drove a 1967 Toytota Corona. My favourite anti-tailgater trick, once I discovered it, was to:
1. Wait for a tailgater to get dangerously close to my bumper and stay there (no point doing this to someone who lost attention for a moment - I think gross negligence should be punsished, but I don't think the net should be cast too wide)
2. While keeping my right foot on the accelerator so as not to lose speed, press the brake pedal with my left enough to turn on the brake lights without actually having any significant braking effect - and waggle the steering wheel rapidly an extremely short distance left and right to make the back of the car sway a little and look like I'm stopping suddenly.
The advantage my trick has over yours is that there's little risk of the tailgating cretin's vehicle coming into contact with yours.
The "I thought I saw a road-hazard" argument was the one I planned on using if ever called on the matter - a small box on the road that must have been blown off as I or the car behind me passed it. You don't want to hit a box - you've got no way of knowing if it's empty or full of bricks until it's too late. Bunnies? Hell, around here we'd get a commendation for taking one of the furry little ringbarkers out. Same with foxes.
It's not that I was travelling slowly - the speedo was reasonably accurate according to the highway speed checkers, so I tended to sit on the limit (exactly mid-way between 60 and 65 was close enough to 100 in metric). I think they just thought "Smallish, older car" and assumed that they should be able to travel faster than I could.
Our plain old (Australian) free-to-air TV blew up a few weeks back, and the cost of repair (with 3 month warranty) was close to the cost of new identical TV (with 2 year warranty). We toyed with the idea of a HDTV, but the repair guy advised us to wait a few years because they still haven't worked out the Australian standard, and we'd risk buying an expensive boat-anchor
Oh, the Australian standard has well-and-truly been worked out.
It's whatever the broadcaster chooses to broadcast as HD
We recently bought a honking-huge LG DLP Rear Projection TV, and I've been taking a look at what qualifies as HD broadcasts - 576p, 720p, 1080 depending on the broadcaster and the phase of the moon. Sometimes, I'll just stick to SD on a smaller set, TYVM, as upscaled tape noise doesn't do much for the viewing experience. For the most part on material specifically produced with HD in mind, though, the difference is phenomenal
The set in question has a built-in HD tuner, does 1280x720 native, and upscales or downscales depending on the broadcast resolution. Yes, that's right... even plain analogue broadcasts get twisted and tweaked and interpolated to display on the 720-line display. And it has a pretty decent analogue tuner too.
Although it would probably be nice to have something that will do 1080p, especially in a few years when the dust settles and the HD disk format war has been fought and won, I don't know that I'd be able to tell the difference.
There's obsolescence, and then there's obsolescence. I don't see my set becoming unusable any time soon. I also don't see me being able to tell the difference between what it can do and full 1080p until 1080p equipment and material becomes common and affordable, at which point a 1080-native set of the same size will probably cost half what I paid for this one, instead of twice what I paid. And in the meantime, I'll have had a 60+ inch TV to watch:-)
That said, with some of the prices on 80cm glowbottle sets right now, unless you wanted a really big screen right now you'd probably be better off buying a CRT and waiting for the next generation of DLP, LCoS, or even LCD or plasma.
>I'm the IT Director for my company here in the northeast US.
I'm the mail admin at a university.
>Our spam percentage over the past year has climbed from about 80% to 91.7% this past month (October 2006)
We only accepted 9.25% of attempts to send mail to our domains in October.
Figures for this year are:
Month Rejected Virus Accepted Total % Accepted Jan 1537406 21956 462832 2022194 22.89% Feb 1570777 11907 532155 2114839 25.16% Mar 1566575 14544 649630 2230749 29.12% Apr 1807829 12659 532450 2352938 22.63% May 2863094 25669 713798 3602561 19.81% Jun 4169771 9100 676440 4855311 13.93% Jul 3424146 13392 624217 4061755 15.37% Aug 2977085 9590 709291 3695966 19.19% Sep 4946817 36077 673710 5656604 11.91% Oct 7752193 24516 792346 8569055 9.25%
Virus figures are pretty low, because many virus sources are (or quickly end up) listed in places like Spamhaus XBL, or are dynamic ranges listed by SORBS or NJABL.
I know RBLs generate a lot of ill-feeling from some people, but given the mail volumes we have to deal with we simply couldn't survive if we didn't have them as a first-approximation. I think part of the problem is that some people see them as set-and-forget, whereas local whitelistings for mail sources that are important to your organisation are extremely important. We use a combination of RBLs for first-approximation, then run SpamAssassin over everything that passes that... we have some locally tweaked scores, and reject at anything that scores higher than 15. Yeah, I know, that looks high... but we have things like Asian webmail that we need to let through, and add things like 5 points for URIBL and RBL tests, even ones like CSMA or PSBL that we wouldn't block against directly. Overall, it works reasonably well.
Spamassassin, unfortunately, does have false positives, so anything that spamassassin tags gets delivered with a 'possibly spam' header added.
Using the default scores?
I've got the scores tweaked somewhat, and we tag at 5 and reject at 15. RBLs (including ones we wouldn't reject on outright), URIBLs, Razor checks etc are all good for five points each. No PTR gets 10 (although that one's hacked together in our mimedefang filter). Try skewing the scores somewhat for tests that look super-spammy, and reject at some relatively high threshold that requires a few to be tripped... that seems to limit the SA false-positives to things that would have problems getting through
many other systems anyway.
We have RBL checks (and some mitigating whitelistings) ahead of all that, and find that our system copes really well with even high-volume days. We normally reject around 100000 messages per day, but that's been up to a quarter million on occasions.
Let me just pipe in with my two cents worth. I'm the mail administrator at a University, and have been through all these arguments before.
Rejecting mail outright based on RBL results is an invitation for loads of false positives. If you're getting "1000s of messages per minute," chances are you are a large business or an ISP. Either way, your customers sending mail from foobaz.net (for example) won't be happy when their no longer gets through because some boneheads at spamcop or spamhaus put them on the RBL.
(actually, if they're in SpamCop or SpamHaus, I'd say they probably deserve it; I won't block directly on SORBS-Spam, but I'll cheefully reject on most of the other SORBS lists)
Keep your postmaster address completely unfiltered, and have meaningful information in the rejection notice that invites affected senders to contact you so something can be sorted out. Our mail gateway has lately been rejecting 100000 messages a day (around 80% of attempted deliveries) and I've been getting between four and a dozen reported "false-positives" per month, and we're usually happy to have something temporary in place to let their mail through within a couple of working hours of being informed.
This also brings up the fact that anyone who uses RBLs without knowing what they are or what they might do is grossly irresponsible. Anyone using them (myself included) needs to understand that from time to time an ISP or a trusted business partner might end up in one - and you'll need whitelistings to ensure the straight accept/reject RBL check doesn't apply to those mail sources. This ought to be common knowledge. Anyone who uses this to attack RBLs per se really ought to be attacking a certain administrative mindset that sees RBLs as a magic bullet that doesn't need human oversight - they aren't, and I'm not sure any single technology can really claim to be.
Okay... so we use RBLs, our local shit-list of sources we don't want mail from (to save doing lookups for Charter cable ranges and the like) and local whitelistings. What else do we do here?
We run Sendmail on our external gateways, and use Mimedefang as a milter - that runs messages past ClamAV and SpamAssassin with some locally tweaked rules to increase the scores given by various hits - RBLs we wouldn't dare to block against directly, URIBLs, Razor, etc. We find that a reasonable amount of spam gets caught by this approach, and our first-level approximation based on straight RBL and local shitlist checks keeps the load down to a reasonable level.
If you don't have the proper setup to accept that many messages and do adequate "tag-and-forward" spam analysis,
...and what, store all this stuff in a suspected spam folder, with a whole lot of similarly-scoring legitimate stuff? Tried that... our users were forever losing legitimate mail because they'd assume that everything in there was spam... or they'd get confused, and delete the wrong thing. The recipient never sees it, and the sender doesn't know it wasn't seen... at least if a message is rejected, the sender knows it wasn't seen and has the opportunity to do something about it. If you can keep the amount of accidentally blocked mail really low, and keep inboxes reasonably spam-free, then rejecting does less harm overall than tagging. We let people opt out of the filtering, though... some people really don't get spam, believe it or not.
or at least reject later in the delivery cycle
You have to be kidding... that sounds like a recipe for backscatter.
Of course, you could just set it to randomly drop 80% of all of your incoming email into/dev/null and you would likewise solve your traffic problem.
Now you're just being silly.
In practice, I've found that RBL checks as a first approximation, followed by SA checks on what passes those, keeps the spam level here to being a minor annoy
That was the NTSC version, I assume. Anyone got any comments on the PAL version going through similar gear?
I have the PAL version, and can't say that I've noticed anything wrong with either the cheapie 63cm CRT or the 106cm Widescreen Rear Projection set (three-year-old LG). I'd say that it was significantly better than anything I've ever seen on VHS, and Mrs. Roadkill has quite a good collection of VHS and has been known to order Asian-market DVDs of 80's mini-series that are definitely VHS quality on viewing. I also have that horrible, horrible HORRIBLE 4:3 transfer of Monty Python and the Holy Grail, so I know what VHS-quality DVDs look like.
I guess what I'd like to know is, what does it look like if the set is fed from an interlaced, standard-def output? Is this some weird trick in how the material is mastered to make it look like crap when upscaled for a HD display device, and thus serve as a great in-store side-by-side comparison for selling the new format? More importantly, if that's what happened, can we expect more of that kind of bullshit with future DVD releases?
Some time in the next five years or so I'll probably end up buying a high-def display. I'll probably also end up buying a new high-def disc player to minimise the amount of clutter and keep the loungeroom reasonably kid- and wife-friendly. I will NOT be happy if I find that half the titles I buy between now and then are mastered in a manner that makes them look like crap when played back through my new gear.
Organising a nice large lump of rock and ice or some interesting exotic matter to intersect with the planet in question every couple of hundred million years ought to keep the infestations down to the point where it's economically viable, without needing to leave a caretaker.
I'm just waiting for them to get really pissed when they find out that their last sanitizing attempt at Tunguska failed so dismally and we've squandered their resources.
Bush hasn't been president since the '80s, so that cannot be true!
We are at war with Eurasia. We have always been at war with Eurasia.The chocolate ration has always been at its present level. Officials from Miniluv will be around to help you shortly.
Slashdot Mirror
I want to know when I can get a shirt made out of it.
Both you and grasshoppa are right... for the subsets of the spam problem that you have to address.
My credentials? I run the mailfilters at a university with +50k student addresses, and around 3000 staff addresses. We typically reject a couple of hundred thousand messages daily. So, while our situation would probably turn grasshoppa's hair grey, it's a drop in the bucket compared to yours.
Spam is reasonably surmountable provided you have enough resources to throw at it. However, sometimes that doesn't scale too well
For example, I have two external relays load-balanced and ticking away running a whole lot of RBL checks, SpamAssassin checks and even using the Sane Security add-on signatures for CLAM while they do the regular virus scans. They presently have a whole lot of headroom, and if that ever diminished I could probably get a few thousand dollars for another box to share the load without too many questions given how well our solution works for us and how little it is costing otherwise. I imagine grasshoppa would have similar success with his or her solution. If you're talking hundreds of thosands of messages per hour, scaling my approach out to process that kind of message volume on a corporate budget dictated by the returns your bosses require could be interesting to say the least.
I'd add another item to the list, which some might see as a part of "No Magic Bullet", but I see it as sufficiently important to warrant its own entry - "There is no one-size-fits-all solution" . What works well-enough for you or me or grasshoppa will probably work well enough for similiarly sized organisations with similar resources and similar threats and similar values of "work well enough" - change any of that, though, and it's a whole new problem. The spammers are probably laughing their arses off over the kind of kind of religious wars people have over whose spamfiltering solution is best (it's mine, by the way, but then I also use vi to edit my conf files so of course it would be). And diversity of solutions is probably a good thing, as it means the spammers need to work harder to get past a whole lot of different approaches to get their garbage out.
Has anyone reverse-engineered the watermarks sufficiently to enable someone to add bogus watermarks to printouts from non-watermarking printers? That could be fun, once RFID is sufficiently commonplace - walk past a stack of printers in a store, capture their serial numbers and give them to your anarc^bastard friends on the other side of the country who then procede to create bogus watermarks that get used to send threatening letters over the next year or two and get law enforcement hassling innocent people whose only crime was to buy a printer and send in the warranty card.
(and who the hell modded me OT? Did they actually RTFA? And do they still have enough modpoints to come back and mod this "Flamebait"?)
Am I the only one who's annoyed by bar codes on CD covers and books?
Of course, this probably wouldn't fare too well on a re-issue of the White Album...
Of course, we may have some areas we agree on. If someone wanted to modify their display drivers to allow them to play HDCP content through their own capable but ageing projector or a DVI-only flat-panel, I see no reason why they shouldn't be able to do so. If someone wanted to modify their wifi drivers in a way that gained them some advantage but did so to the detriment of their neighbours, they ought not consider it in the first place - but because some people will, I have no problem with things that make it difficult for them to do so. Some people simply don't give a shit about what effect their actions will have on others, and so we need laws and rules and maybe the occasional piece of mob justice or political activism.
For some things, I am. For me, it's a matter of using a suitable tool for the task at hand. Don't I have the freedom to choose the tool that fits my circumstances best, even if I have to pay for it? As someone else got modded troll for saying, it sometimes seems like the FSF wants everyone to have the freedom to do things the way the FSF wants them to be done. And if I dislike something that a software vendor is doing, I have the freedom to look at alternatives and vote with my dollars - and even to write something that suits my needs better than some commercial offering and release it under whatever licence conditions I like.Additionally, for some devices with binary-only drivers, there are other matters that prevent the manufacturers opening up the drivers fully. The one that everyone always brings up first is that of the hardware manufacturers not always owning all the IP that goes into the drivers, and thus not being able to release all the specs. That's perhaps a very real and legitimate concern, and ought to be respected by people who get all pissy when some piece of free code gets misappropriated. The one I'd like to focus on though, even if it's slightly off-topic given that the discussion is video drivers, is that of open-source drivers not necessarily being compliant with regulations conceringing the operation of a particular piece of hardware. I've have EXTREMELY good results from an LT Winmodem (from the junkbox) I've dropped into a smoothwall box (also from the junkbox) that I built to save a relative from pr0ndiallers, but I would not like to see completely open-source drivers for such devices. "Why not?" you ask, quite legitimately. Well, who is most qualified to write drivers that affect things like how strong a signal and what encoding to pump down the phone line - some random hacker, or someone who knows PSTN technology? And who is liable if a home-built driver causes the modem to do nasty things to the phone line? At least with a vendor-supplied driver you can point the finger back to them, but here in Oz with a homebrew driver you could be facing a $12K fine if your softmodem does something funky to the exchange. With some wireless chipsets where a lot of stuff is done in software the same arguments could apply - it would theoretically be possible for people to write their own drivers that push the gear out-of-spec, which may be fine if you're out in the middle of the desert but which is definitely not desirable if you have to share the spectrum with your neighbours.
Given my 'druthers, I'druther have hardware that just does what it's asked to do with minimal CPU effort required - "Here are these bits, you know what to do". But then, I do recognise that software-based control of those devices means that they can be easier to update - for example, installing new drivers rather than flashing and hoping that you don't brick the device. I also recognise that CPUs are fast enough these days for the additional load from these kinds of devices to be neglible compared to that from the eye-candy most people have on their desktops these days, so the old system performance arguments usually don't hold water any more. Finally, from a freedom perspective, what's the difference between a black-box that you plug into your computer and a black box that you run on your computer? You don't hear people complaining that they're somehow oppressed because they can't get the source code for their external throw-bits-at-it-down-the-serial-line modem, yet somehow a software modem or wireless nic with binary-only drivers is the spawn of satan?
I don't care - mod me down if you feel compelled to do so. Honestly, though, I can't find a reason to get my knickers in a twist about binary-only drivers and their supposed deleterious effects on free software. If a manufacturer is prepared to put together good, stable and functional binary drivers, then good for them - they'll at least be considered next time I need a piece of hardware. If it gives me the freedom to choose from a wider range of products at a wider range of prices, I'm all for it. If you're concerned with the gear getting orphaned, think about the last time you used a piece of older gear - and what you used it in. Sure, I can p
In two years, Microsoft will buy CodeWeavers. Each MS application will ship with the then-current version of CrossOver, for Linux and MacOS. Some time after that, Microsoft will do a premium distro that includes five years worth of service packs and CrossOver updates and their own desktop, so there's still scope for MS to do OEM and retail deals and continue to own the OEM space. Those who want to run MS apps under their favourite distro or even a roll-your-own installation can do so with a compatibility layer sanctioned and produced by Microsoft, said layer being paid for by the increased market share Microsoft gain and MS's reduced costs from being able to leverage outside community development. Okay, I kind of gagged on that last bit too, but I can see MS putting effort into fixing various kernel and system software issues and doing performance tweaking so their software works better, and under the GPL they'd need to give that stuff back. Their desktop they could licence commercially, of course, but it'd be in their interests to make sure anything that made their software work better on Linux got distributed as widely as possible. If MS got behind allowing their apps to run on Linux, that could only be a good thing for Linux in general.
So, where's my crack-pipe?
There are a number of smaller businesses out there with something like Mercury32 or MDaemon running on Win95/98/ME, with halfway-decent firewalls that keep the bad guys from attacking directly or no attackable services running, and no web browsing from those boxes to expose them to the various web-based exploits that affect the out-of-date browsers on their machines. These set-ups were probably installed years ago by various consultants, and have been left alone because they continue to do what they're meant to. It's possible for something to have a Win98/98/ME fingerprint, legitimately send mail directly to your servers and be no threat to your or your users.
Okay, here's the REAL deal.
Harrison Ford took the Indiana Jones role over the Han Solo one because it's going to be a much better movie.
The rumour that it takes place in the sixties is true, and fits in nicely with the Mr Ford's present age.
What hasn't been widely revealed is that Sean Connory *will* be in the movie, although the role will probably surprise many people.
Since Satan owns the pink slip for the soul of pretty much everyone who has ever worked in motion pictures, he can shuffle the deck however he sees fit... and some interesting studio mergers mean that Sean Connory will play an elderly James Bond who fell through a temporal rift as the result of Xindi interference with Earth history - the theory being that if they could get all the kids hooked on beer and acid and dope then warp drive would never be invented. Little did they realise that Optimus Prime would ride in on My Little Pony and save the day by assassinating Kennedy and illegitimately fathering Rosie ODonnell with, you guessed it, Rosie ODonnell - who fell through the same temporal rift James Bond fell through. Pygmies re-discover left-over gou'auld technology that permits them to build hypersonic blow-dart weapons, which are capable of destroying ICBMs and thereby save the USA from the tyranny of total destruction when they decide to make the Ukraine glow in the dark...which happens two-thirds of the way through the movie, because the Ark of the Covenant (which was stolen from Area 51 by the Xindi) has been given to the Russians, who are using it to try to re-animate a cut-n-shunt SuperPolitician they've made from the cryogenically preserved remains of Adolph Hitler, Josef Stalin and Walt Disney - but exposure to nuclear fallout causes this re-animated monstrosity to sprout wings and fly to Tokyo, where as Mothra it does battle with Godzilla until Indiana Jones...
Sorry, I've given too much away already. You'll just have to buy a ticket like everybody else.
Its picture quality was far superior to the VHS recorder we had at the time, but I have to wonder if that was due to the fact that it clearly wasn't as "built to a price" as the VHS units we could afford - although that could have been in part due to the fact that the bank had used high-grade tapes and the only things I could buy new for it were high-grade, and I wasn't about to spend $10 on a VHS tape for one-off viewing when I could get three lesser-grade ones for the same price. I wonder how much of the perceived quality difference between VHS and Beta has been shaped by the quality of the equipment and media used in the comparisons? Lounge rooms are hardly a laboratory environment, and it's reasonable to expect a mid-eighties Sony beta recorder with good tapes to outperform a 1990 Korean VHS cheapie with budget tapes.
That said, I'd been considering removing ORDB from our checks for some time. On days when NJABL and SpamHaus were picking up 30-50k messages each, ORDB would pick up between one and five messages. So, although it's sad to see their passing, for me at least they weren't that important a part of my spam-minimisation strategy.
On a serious note, though, it would be very interesting to see some information on the kind of things that drive people into that kind of activity. I can't honestly say what I would or would not do if faced with extreme poverty and a low-risk data-entry position like that.
About twelve years ago, I drove a 1967 Toytota Corona. My favourite anti-tailgater trick, once I discovered it, was to:
1. Wait for a tailgater to get dangerously close to my bumper and stay there (no point doing this to someone who lost attention for a moment - I think gross negligence should be punsished, but I don't think the net should be cast too wide)
2. While keeping my right foot on the accelerator so as not to lose speed, press the brake pedal with my left enough to turn on the brake lights without actually having any significant braking effect - and waggle the steering wheel rapidly an extremely short distance left and right to make the back of the car sway a little and look like I'm stopping suddenly.
The advantage my trick has over yours is that there's little risk of the tailgating cretin's vehicle coming into contact with yours.
The "I thought I saw a road-hazard" argument was the one I planned on using if ever called on the matter - a small box on the road that must have been blown off as I or the car behind me passed it. You don't want to hit a box - you've got no way of knowing if it's empty or full of bricks until it's too late. Bunnies? Hell, around here we'd get a commendation for taking one of the furry little ringbarkers out. Same with foxes.
It's not that I was travelling slowly - the speedo was reasonably accurate according to the highway speed checkers, so I tended to sit on the limit (exactly mid-way between 60 and 65 was close enough to 100 in metric). I think they just thought "Smallish, older car" and assumed that they should be able to travel faster than I could.
It's whatever the broadcaster chooses to broadcast as HD
We recently bought a honking-huge LG DLP Rear Projection TV, and I've been taking a look at what qualifies as HD broadcasts - 576p, 720p, 1080 depending on the broadcaster and the phase of the moon. Sometimes, I'll just stick to SD on a smaller set, TYVM, as upscaled tape noise doesn't do much for the viewing experience. For the most part on material specifically produced with HD in mind, though, the difference is phenomenal
The set in question has a built-in HD tuner, does 1280x720 native, and upscales or downscales depending on the broadcast resolution. Yes, that's right... even plain analogue broadcasts get twisted and tweaked and interpolated to display on the 720-line display. And it has a pretty decent analogue tuner too.
Although it would probably be nice to have something that will do 1080p, especially in a few years when the dust settles and the HD disk format war has been fought and won, I don't know that I'd be able to tell the difference.
There's obsolescence, and then there's obsolescence. I don't see my set becoming unusable any time soon. I also don't see me being able to tell the difference between what it can do and full 1080p until 1080p equipment and material becomes common and affordable, at which point a 1080-native set of the same size will probably cost half what I paid for this one, instead of twice what I paid. And in the meantime, I'll have had a 60+ inch TV to watch :-)
That said, with some of the prices on 80cm glowbottle sets right now, unless you wanted a really big screen right now you'd probably be better off buying a CRT and waiting for the next generation of DLP, LCoS, or even LCD or plasma.
>I'm the IT Director for my company here in the northeast US.
I'm the mail admin at a university.
>Our spam percentage over the past year has climbed from about 80% to 91.7% this past month (October 2006)
We only accepted 9.25% of attempts to send mail to our domains in October.
Figures for this year are:
Month Rejected Virus Accepted Total % Accepted
Jan 1537406 21956 462832 2022194 22.89%
Feb 1570777 11907 532155 2114839 25.16%
Mar 1566575 14544 649630 2230749 29.12%
Apr 1807829 12659 532450 2352938 22.63%
May 2863094 25669 713798 3602561 19.81%
Jun 4169771 9100 676440 4855311 13.93%
Jul 3424146 13392 624217 4061755 15.37%
Aug 2977085 9590 709291 3695966 19.19%
Sep 4946817 36077 673710 5656604 11.91%
Oct 7752193 24516 792346 8569055 9.25%
Virus figures are pretty low, because many virus sources are (or quickly end up) listed in places like Spamhaus XBL, or are dynamic ranges listed by SORBS or NJABL.
I know RBLs generate a lot of ill-feeling from some people, but given the mail volumes we have to deal with we simply couldn't survive if we didn't have them as a first-approximation. I think part of the problem is that some people see them as set-and-forget, whereas local whitelistings for mail sources that are important to your organisation are extremely important. We use a combination of RBLs for first-approximation, then run SpamAssassin over everything that passes that... we have some locally tweaked scores, and reject at anything that scores higher than 15. Yeah, I know, that looks high... but we have things like Asian webmail that we need to let through, and add things like 5 points for URIBL and RBL tests, even ones like CSMA or PSBL that we wouldn't block against directly. Overall, it works reasonably well.
I've got the scores tweaked somewhat, and we tag at 5 and reject at 15. RBLs (including ones we wouldn't reject on outright), URIBLs, Razor checks etc are all good for five points each. No PTR gets 10 (although that one's hacked together in our mimedefang filter). Try skewing the scores somewhat for tests that look super-spammy, and reject at some relatively high threshold that requires a few to be tripped... that seems to limit the SA false-positives to things that would have problems getting through many other systems anyway.
We have RBL checks (and some mitigating whitelistings) ahead of all that, and find that our system copes really well with even high-volume days. We normally reject around 100000 messages per day, but that's been up to a quarter million on occasions.
(actually, if they're in SpamCop or SpamHaus, I'd say they probably deserve it; I won't block directly on SORBS-Spam, but I'll cheefully reject on most of the other SORBS lists)
Keep your postmaster address completely unfiltered, and have meaningful information in the rejection notice that invites affected senders to contact you so something can be sorted out. Our mail gateway has lately been rejecting 100000 messages a day (around 80% of attempted deliveries) and I've been getting between four and a dozen reported "false-positives" per month, and we're usually happy to have something temporary in place to let their mail through within a couple of working hours of being informed.
This also brings up the fact that anyone who uses RBLs without knowing what they are or what they might do is grossly irresponsible. Anyone using them (myself included) needs to understand that from time to time an ISP or a trusted business partner might end up in one - and you'll need whitelistings to ensure the straight accept/reject RBL check doesn't apply to those mail sources. This ought to be common knowledge. Anyone who uses this to attack RBLs per se really ought to be attacking a certain administrative mindset that sees RBLs as a magic bullet that doesn't need human oversight - they aren't, and I'm not sure any single technology can really claim to be.
Okay... so we use RBLs, our local shit-list of sources we don't want mail from (to save doing lookups for Charter cable ranges and the like) and local whitelistings. What else do we do here?
We run Sendmail on our external gateways, and use Mimedefang as a milter - that runs messages past ClamAV and SpamAssassin with some locally tweaked rules to increase the scores given by various hits - RBLs we wouldn't dare to block against directly, URIBLs, Razor, etc. We find that a reasonable amount of spam gets caught by this approach, and our first-level approximation based on straight RBL and local shitlist checks keeps the load down to a reasonable level.
You have to be kidding... that sounds like a recipe for backscatter.
Now you're just being silly.
In practice, I've found that RBL checks as a first approximation, followed by SA checks on what passes those, keeps the spam level here to being a minor annoy
No I'm not.
Yes, we are.
Shutupshutupshutupshutupshutup!
I have the PAL version, and can't say that I've noticed anything wrong with either the cheapie 63cm CRT or the 106cm Widescreen Rear Projection set (three-year-old LG). I'd say that it was significantly better than anything I've ever seen on VHS, and Mrs. Roadkill has quite a good collection of VHS and has been known to order Asian-market DVDs of 80's mini-series that are definitely VHS quality on viewing. I also have that horrible, horrible HORRIBLE 4:3 transfer of Monty Python and the Holy Grail, so I know what VHS-quality DVDs look like.
I guess what I'd like to know is, what does it look like if the set is fed from an interlaced, standard-def output? Is this some weird trick in how the material is mastered to make it look like crap when upscaled for a HD display device, and thus serve as a great in-store side-by-side comparison for selling the new format? More importantly, if that's what happened, can we expect more of that kind of bullshit with future DVD releases?
Some time in the next five years or so I'll probably end up buying a high-def display. I'll probably also end up buying a new high-def disc player to minimise the amount of clutter and keep the loungeroom reasonably kid- and wife-friendly. I will NOT be happy if I find that half the titles I buy between now and then are mastered in a manner that makes them look like crap when played back through my new gear.
/dev/altimiter not found /dev/wing/left - printer on fire?
GE-xxxx: scsi2: AEN: WARNING: SMART threshold exceeded: Engine #3
Kernel panic: defect on
I'm just waiting for them to get really pissed when they find out that their last sanitizing attempt at Tunguska failed so dismally and we've squandered their resources.