That would seem to indicate that if the ISS didnt make any prograde burns, it too would be entering the atmosphere in several weeks.
According to Wikipedia, "The station, on average, loses 100 meters of altitude per day." (and is periodically reorbitted).
Of course the station has much higher mass compared to the suit. Since (for a given exposed surface area) the atmospheric drag is the same but the mass is less for the suit, the drag will decellerate the suit at a higher rate to the station. Obviously the surface areas aren't the same, but the station is waaaay heavier. As the orbitting object slows down, it loses altitude and drops into thicker atmosphere so a reasonably small difference in the drag:mass ratio to start with leads to an amplified effect as time goes on.
Come to think of it. - Every force has an equal and opposite. which indicates that the ISS crew are going to have to fashion some sort of rocket to get the suit going.
Again, the mass of the station is hugely more than the mass of the suit so the effect on the station is tiny. Besides, does it matter if you alter the station's velocity slightly? It makes sense to throw the suit towards the "back" of the station since otherwise you might catch up with it as it slows down - throwing the suit backwards increases the station's orbital velocity slightly, increasing the altitude of the orbit (which simply means you leave it a bit longer before you have to reboost the station's orbit). And you don't need to worry about upsetting the station's attitude since it has gyroscopes to maintain that.
On a related note, read up on the use of gyroscopes for attitude control on space craft - it's very interesting. The gyroscopes provide a "fixed" object to rotate the spacecraft around, so you can maintain the attitude of the craft by simply rotating the craft around the gyroscopes with electric motors instead of having to fire reaction control thrusters (which uses valuable fuel that you have to fly up from ground).
Because my companies policy is to force users to change passwords every month or so AND the passwords must never be duplicated, I simply pick the date that I change it in the format YYYYMMDD
This is why forcing regular password changes is a bad idea and doesn't do anything to help security. Far better is to just enforce a strong password.
If someone gets your password, you really think they'll wait a couple of months before using it? No - they will use it immediately. I.e. before you've been forced to do your monthly password change.
When you think about it, a crowbar can be used to defeat pretty much any security scheme.
A crowbar can only usually defeat the server security if you break into the server room... which is (hopefully) going to be harder than breaking into some random employee's home and taking the post-it off the monitor they're using when teleworking.
Just as long as they're being appropriately hidden.
There is something to be said for a report like Microsoft's, which has proper reasoning behind it, etc. But NetGear's idea of telling the average end-user that "the experts are wrong, there's no problem writing your password down" just encourages people to write their laptop password on a post-it and stick it to their laptop (which is *always* a stupid thing to do).
If you're going to tell people to do something that may risk security, you _must_ tell them when it's appropriate and how to limit the security risk.
That concept includes all your basic applications too (browser, media player, burning/ripping tools, etc)
How do you define what the "basic" application is? To a web surfer, the browser is a basic application, to a photographer photoshop is a basic application.
I think your arguement about everyone wanting a free(dom) OS is flawed - techies want a free OS, most non-techies don't care since they don't have the skills to use that freedom anyway.
Personally I don't see a problem with paying for a closed source application if you need it, so long as your data is stored in an open format. Of course, in my experience, FOSS software is usually great for most jobs and I have no real need to buy closed software. And from my techie point of view, I prefer FOSS because I can fix bugs and hack in new features myself, but for the average user this is a non-issue.
Was it ever, at least 6809 - m68k was understandable by mere mortals and had a certain elegance.
Quite. But whichever way you look at it, we shouldn't need to have the hardware bloat of remaining compatable with 20 year old hardware. I think most people using FOSS systems have no problem with adopting a completely incompatable, but much better architecture - it's only the people tied to closed systems who need the backwards compatability.
MS users just don't do that - not more than once or twice, anyway.
I'm afraid in my experience, MS users *do* do it over and over and get burnt over and over. (Strangely reminiscent of the Simpsons episode "Duffless" where Lisa wires a cupcake to give Bart a shock, but he keeps reaching for it anyway, never learning)...
Diversity is a useful thing of itself, an x86 monoculture can't be good for us long term.
This is a significant point. IMHO, if an open solution like Linux had taken the "market lead (for desktops)" back in the 80386 days, noone would be using x86 hardware anymore. x86 hardware is really not the best design (any more), the only reason we keep using hardware that's backward compatable with x86 is because the majority of people are using closed software that _can't_ be recompiled for a different architecture. And of course, with the majority of people needing x86 hardware, the price of x86 kit is driven down making it unfeasable to go with a "better" architecture.
I think a good example of the sort of market we would get if the desktop scene wasn't dominated by a closed system that only runs on one architecture is the PDA sector: PDA manufacturers aren't tied into a specific architecture to the same extent as PC users - if some new hardware comes along that has a significant advantage over the old hardware they will happilly dump the old architecture in favor of the new one.
When those pulses go dark, the wires are still around routing telephone calls.
Like it or not, the phone network is steadilly vanishing in favor or VoIP solutions. Yeah, ok - the vast majority of calls are currently going over the PSTN but I'm not sure it's going to be _that_ long before this changes (I should think that in 10 - 20 years the majority of calls will be over the Internet).
Not to mention the fact that the PSTN _is_ connected to the Internet - your DSL connection, etc runs over the same digital network as your phone calls so there are physical connections between the two. Having physical connections between the networks does mean there is a posibility (however remote) that an attack on one could affect the other.
Though the paper-less office has been a longstanding goal, it is totally a dream. Everything has a papertrail and can be backed up.
This definately depends on what work you're doing. Certainly in my job, if the Internet connection dies then work (almost) completely stops for the duration of the outage.
I can always testify to the fact that the majority of sysadmins I deal with are lazy and/or stupid - I'd estimate that about 90% of the hard drives that die in customers' servers are not backed up because the customer didn't consider any disaster scenarios or was too lazy/busy to actually sort out a backup solution. The fact that data _can_ be backed up is a moot point - when your data gets wiped out the only thing you need to consider is whether the data _is_ backed up.
People compare September 11 with a lot of things, but comparing it with a crack-fest?
I don't think they are comparing the 9/11 attacks themselves to a crack-fest, they are compairing the resulting economic disruption to something that could be done through a coordinated cracking session. I'm not wholly convinced that economic disruption of such large proportions can be coordinated through cracking though.
Don't use Windows, use OS's designed with security in mind.
I'd agree with this - certainly for mission-critical systems anyway. However, *all* OSes must be kept patched and up to date - a 4 year old Linux distribution is probably just as vulnerable as a 4 year old Windows release, it's only when you keep them patched up to date that Linux gets significantly more secure than Windows.
For workstations, Windows is sometimes a necessary evil but I think in most cases you *can* ditch Windows in favor of a better OS (Linux or consider OS X if Linux won't run the software you need).
Use SELinux or equivalent on mission critical nodes.
SELinux is still far from perfect on current distributions - certainly under Fedora Core 3 the supplied policies are too restrictive in a number of cases (Apache can't do a lot of stuff you want it to do, etc.). Whilest you _could_ rewrite the SELinux policies, you probably need a brain the size of a planet.:)
And secondly, educate the users and gain a culture of safety.
This is probably _the_ most important point. No matter how much you secure the software, the users are always a weak point. For the servers this isn't a big deal coz anyone who can log into them has (hopefully) got a clue. But you don't need to compromise the servers to cause disruption - once a single workstation has been compromised (maybe the user wanted to look at the cool new screensaver someone mailed them, whcih turned out to be a trojan) then your network is unsafe - your firewall won't do you much good now.
Your value for Boltzmann's constant is way too high.
Hmm, ok, I got the figure from Wikipedia which recons you multiply the energy in eV by 11,605 to get the temperature in K. Is that wrong or am I misunderstanding it somehow?
Also, temperature isn't necesarily well-defined or meaningful at these very low densities (that's right, density, screw pressure). Heat content might be a better way to look at things.
Well, that was sort of my point that temperature is probably not such a useful thing to measure at such low densities.
I don't think 'hot' is used to describe a location that is 7 billion miles from the sun
Actually, "hot" (or temperature) is really describing the energy of the particles in the area. Inside the solar system, the solar wind is moving at pretty high speeds - wikipedia suggests energies of 500 KeV. Using the Boltzmann Constant we get 500,000 x 11,605 = 5.8 billion degrees K (Sounds a lot - can some astrophysacists check my figures please:).
Once you get to the termination shock, the solar wind is moving at much slower sub-sonic speeds. Not sure what energies we're talking about here but they're going to be a *lot* lower... A bit of googling suggests He energies somewhere around the 5.2 KeV area (5,200 x 11,605 = 60 million degrees K).
Of course, although the matter may be "hot", there isn't much of it - the low density of matter means that there isn't much "heat" (compare - a cigarette is "hot" (it's gonna burn you) whereas a central heating radiator is not as hot but generates more "heat" (it'll warm your room better than the cigarette because it's total energy output is much greater, even though it's temperature is less)).
Disclaimer: IANAAP (Astro-Physacist) so the above could be crap, but that is how I understand it.
Of course, admins love Windows. They hate Linux and Macs, because companies running these systems don't need as many admins.
I was installing some (Linux based) networking kit at a reasonably big customer a few weeks ago.
Now, from my perspective, if someone told me that one of my Linux systems had been compromised I'd say "oh shit" and be dropping everything to take it offline and reinstall, or at least clean it ASAP. (Yep, compromises _do_ happen on Linux kit but are reasonably infrequent if you keep stuff up to date, and 9 times out of 10 are caused my human error such as forgetting to turn on scheduled yum updates or leaving a weak password on a server).
While I was doing this install, I had to inform the customer that it looked like one of his (many) windows servers had been hit by some worm (lots of network traffic hitting random addresses on the internet on obviously microsofty ports). The reaction kind of shocked me - it was pretty much "oh well, happens all the time, I'll fix it later".
I think this says a lot about Windows admins - they have so many security problems that a random exploit (which, for all they know, could be posting confidential data all over the place) is considered pretty much the norm and unimportant.
With a helicopter, for instance, as long as you "float" over the deck without hitting anything, you can land anywhere.
I'm guessing that if the deck is going up and down by 20 feet every few seconds then "floating" over the deck without hitting it gets somewhat tricky. I know nothing of actually landing choppers in such conditions, but I understand that being lowered from a chopper onto the deck can be very dangerous, leading to broken bones or even death (large waves really do lift ships reasonable distances at quite a speed).
Remember when you started to pay cable TV to get rid of the commercials?? remember how did you feel when your cable company started to put commercials in their channels?
Errm, I dunno about you, but I don't pay Google...
I think we need to face the fact that Google now exists primarily to sell ads.
Is this not how it's always been? The primary reason for a company to exist is to make money. Google makes all it's money through ads, so obviously Google's primary objective is to sell ad space. Ad space is more valuable if more people see it, so if google continue to provide an excellent service then they will attract more people and make more money from the ads - everyone wins.
Right now, there are a zillion anonymous proxies on wired connections.
It's actually pretty hard to _guarantee_ you are anonymous on the internet. If you use an anonymous proxy then your IP will be hidden from the end web server you are contacting, but there is *no way* to know if the anonymous proxy is keeping logs. The authorities can track your web accesses back to the proxy, and if the proxy is keeping logs then it's very easy for the authorities to get the logs through a court order and tie the web server accesses to your real IP address.
The only way to guarantee an anonymiser isn't holding incriminating evidence against you is if you administer it yourself... and in that case you're probably screwed anyway.
I don't understand this idea that poor UI design is "OK for techies." We all benefit from good design, even power users.
Yes, techies benefit from good UI design too. However, techies can get by with an over-complex and inconsistent UI whereas non-techies would probably just get confused, frustrated and give up.
To most techies, having a feature-rich application is often more important than having a really simple UI, whereas to most non-techies they are happy with not so many features and a really simple UI.
Of course, "feature-rich" and "really simple" don't always (often?) go together very well. I think the current trend in Gnome is a perfect example of this - they're ripping out all the useful functionality because they claim it complicates things too much. (I used to use Gnome exclusively... then it lost all the features I liked so I stopped).
The thing that frustrates me more than anything else is that there's no excuse for this, except for the choices made by the project leader.
I don't think this is down to a choice my Isaac but simply down to the ad-hoc nature with which FOSS projects such as Myth are developed. Different bits come from different people and so consistency can be limited. What they actually need is a "UI guru" or a organised team of UI people to sanitise the UI and new elements as they are implemented, however that involves interested parties to have time to do that sort of thing and many (techies) perceive the UI as somewhat secondary to the "cool stuff" (I am often no different with projects I'm working on, preferring to play with cool new stuff instead of boring sanitisation of other people's work).
There's a lot to be read into the fact that users will now be able to vote on which features should be prioritized for development.
I must make a very important point here that many people (users) just don't get: with FOSS software, everyone developing it is freely putting their time into the project and getting nothing in return except the features they are implementing. This means that in general, developers working on a FOSS project will only implement features that they themselves want. So basically, if you want something doing then you may well have to do it yourself, you certainly shouldn't expect someone else to give up their time to do it.
Commercial projects, OTOH, are not bound by such problems and so a developer may well develop a feature you want. However, this only works to a point - larger companies (i.e. those who don't significantly benefit from a minority group buying a product) are unlikely to implement features for individual users and in many cases get too much input from the marketting people who want "cool" stuff people are never going to use or want to rip out functionality on the grounds that it "overcomplicates" it. When was the last time Microsoft implemented a feature you asked for?:)
I think there may well be an advantage in FOSS projects which have commercial backing so long as the commercial side doesn't have _too much_ influence over the featureset, etc (Asterisk seems to be a good example of an FOSS project with commercial backing)
Oh yes, I know I shouldn't reply to myself but a point I missed:
There are no CAMs available for decoding Sky channels, so you have to use a normal Sky box to decode to analogue and then reencode to MPEG4 instead of just using a DVB-S card to suck the MPEG2 data straight off the satellite dish. This sucks but I don't think Ofcom (or whoever) is likely to force Sky to sell a CAM, which gives Sky+ a bit of an advantage.:(
Gotta disagree. Myth is nice but is still FAR lacking in many ways - UI and ease of development in particular (speaking from some experience).
I find Myth pretty feature-rich, and it certainly seems to screw up and forget to record stuff less frequently than the Sky+ boxes some of my friends have.
You might be right about the UI to some extant - it's mostly ok for the techie but probably not so suitable for the general public (but then are the general public going to build their own Myth box or just buy one of the commercial PVRs?).
I think the main problem with the UI from my point of view is the recording priorities stuff - I don't like having to juggle integer priorities for all my programs and would prefer to just see a list of shows ordered by priority and be able to move a show up and down the list.
There is also some inconsistency with key bindings too - most of the UI looks in the key bindings database to find out which key is "select", whcih is "play", etc. However, some parts of the UI make assumptions instead - i.e. expecting Enter to be "select". But that's reasonably minor and probably doesn't affect most people.
I've not really done any UI development for Myth (just added a few controls to some of the setup screens...), although I did write some of the back end code (A/V synchronisation routines, etc) and can't say it was that hard to implement, despite not really being a C++ coder - I usually just use C so there was a slight learning curve there.
I like Myth, but it has many warts, and missing program guide data is not one of them.
I use the RadioTimes listings and I have to say that everything has got a *lot* better since RT started providing machine readable listings - the site scraper used to take hours and every so often they'd change something that broke it. There is still the occasional problem that programmes which are rerun several times during the 2 week period you get listings for sometimes don't have matching descriptions or subtitles so you get 2 recordings but for the most part it's not bad. Of course I'd like radio listings too (used to get them from the scraper but they don't provide machine readable radio listings).
Slashdot Mirror
That would seem to indicate that if the ISS didnt make any prograde burns, it too would be entering the atmosphere in several weeks.
According to Wikipedia, "The station, on average, loses 100 meters of altitude per day." (and is periodically reorbitted).
Of course the station has much higher mass compared to the suit. Since (for a given exposed surface area) the atmospheric drag is the same but the mass is less for the suit, the drag will decellerate the suit at a higher rate to the station. Obviously the surface areas aren't the same, but the station is waaaay heavier. As the orbitting object slows down, it loses altitude and drops into thicker atmosphere so a reasonably small difference in the drag:mass ratio to start with leads to an amplified effect as time goes on.
Come to think of it. - Every force has an equal and opposite. which indicates that the ISS crew are going to have to fashion some sort of rocket to get the suit going.
Again, the mass of the station is hugely more than the mass of the suit so the effect on the station is tiny. Besides, does it matter if you alter the station's velocity slightly? It makes sense to throw the suit towards the "back" of the station since otherwise you might catch up with it as it slows down - throwing the suit backwards increases the station's orbital velocity slightly, increasing the altitude of the orbit (which simply means you leave it a bit longer before you have to reboost the station's orbit). And you don't need to worry about upsetting the station's attitude since it has gyroscopes to maintain that.
On a related note, read up on the use of gyroscopes for attitude control on space craft - it's very interesting. The gyroscopes provide a "fixed" object to rotate the spacecraft around, so you can maintain the attitude of the craft by simply rotating the craft around the gyroscopes with electric motors instead of having to fire reaction control thrusters (which uses valuable fuel that you have to fly up from ground).
Because my companies policy is to force users to change passwords every month or so AND the passwords must never be duplicated, I simply pick the date that I change it in the format YYYYMMDD
This is why forcing regular password changes is a bad idea and doesn't do anything to help security. Far better is to just enforce a strong password.
If someone gets your password, you really think they'll wait a couple of months before using it? No - they will use it immediately. I.e. before you've been forced to do your monthly password change.
When you think about it, a crowbar can be used to defeat pretty much any security scheme.
A crowbar can only usually defeat the server security if you break into the server room... which is (hopefully) going to be harder than breaking into some random employee's home and taking the post-it off the monitor they're using when teleworking.
Just as long as they're being appropriately hidden.
There is something to be said for a report like Microsoft's, which has proper reasoning behind it, etc. But NetGear's idea of telling the average end-user that "the experts are wrong, there's no problem writing your password down" just encourages people to write their laptop password on a post-it and stick it to their laptop (which is *always* a stupid thing to do).
If you're going to tell people to do something that may risk security, you _must_ tell them when it's appropriate and how to limit the security risk.
can anyone recommend a centralized password storage software solution that works well for them?
:)
Store 'em on usenet...
But I do think $2 a gallon is pretty outrageous...
Here in the UK I pay about 85p/litre ATM... I'm sorry but fuel in the US is _cheap_.
That concept includes all your basic applications too (browser, media player, burning/ripping tools, etc)
How do you define what the "basic" application is? To a web surfer, the browser is a basic application, to a photographer photoshop is a basic application.
I think your arguement about everyone wanting a free(dom) OS is flawed - techies want a free OS, most non-techies don't care since they don't have the skills to use that freedom anyway.
Personally I don't see a problem with paying for a closed source application if you need it, so long as your data is stored in an open format. Of course, in my experience, FOSS software is usually great for most jobs and I have no real need to buy closed software. And from my techie point of view, I prefer FOSS because I can fix bugs and hack in new features myself, but for the average user this is a non-issue.
What, you don't think that proxy server logs things too?
Can you say "SSL"?
Was it ever, at least 6809 - m68k was understandable by mere mortals and had a certain elegance.
Quite. But whichever way you look at it, we shouldn't need to have the hardware bloat of remaining compatable with 20 year old hardware. I think most people using FOSS systems have no problem with adopting a completely incompatable, but much better architecture - it's only the people tied to closed systems who need the backwards compatability.
MS users just don't do that - not more than once or twice, anyway.
I'm afraid in my experience, MS users *do* do it over and over and get burnt over and over. (Strangely reminiscent of the Simpsons episode "Duffless" where Lisa wires a cupcake to give Bart a shock, but he keeps reaching for it anyway, never learning)...
Diversity is a useful thing of itself, an x86 monoculture can't be good for us long term.
This is a significant point. IMHO, if an open solution like Linux had taken the "market lead (for desktops)" back in the 80386 days, noone would be using x86 hardware anymore. x86 hardware is really not the best design (any more), the only reason we keep using hardware that's backward compatable with x86 is because the majority of people are using closed software that _can't_ be recompiled for a different architecture. And of course, with the majority of people needing x86 hardware, the price of x86 kit is driven down making it unfeasable to go with a "better" architecture.
I think a good example of the sort of market we would get if the desktop scene wasn't dominated by a closed system that only runs on one architecture is the PDA sector: PDA manufacturers aren't tied into a specific architecture to the same extent as PC users - if some new hardware comes along that has a significant advantage over the old hardware they will happilly dump the old architecture in favor of the new one.
When those pulses go dark, the wires are still around routing telephone calls.
Like it or not, the phone network is steadilly vanishing in favor or VoIP solutions. Yeah, ok - the vast majority of calls are currently going over the PSTN but I'm not sure it's going to be _that_ long before this changes (I should think that in 10 - 20 years the majority of calls will be over the Internet).
Not to mention the fact that the PSTN _is_ connected to the Internet - your DSL connection, etc runs over the same digital network as your phone calls so there are physical connections between the two. Having physical connections between the networks does mean there is a posibility (however remote) that an attack on one could affect the other.
Though the paper-less office has been a longstanding goal, it is totally a dream. Everything has a papertrail and can be backed up.
This definately depends on what work you're doing. Certainly in my job, if the Internet connection dies then work (almost) completely stops for the duration of the outage.
I can always testify to the fact that the majority of sysadmins I deal with are lazy and/or stupid - I'd estimate that about 90% of the hard drives that die in customers' servers are not backed up because the customer didn't consider any disaster scenarios or was too lazy/busy to actually sort out a backup solution. The fact that data _can_ be backed up is a moot point - when your data gets wiped out the only thing you need to consider is whether the data _is_ backed up.
People compare September 11 with a lot of things, but comparing it with a crack-fest?
:)
I don't think they are comparing the 9/11 attacks themselves to a crack-fest, they are compairing the resulting economic disruption to something that could be done through a coordinated cracking session. I'm not wholly convinced that economic disruption of such large proportions can be coordinated through cracking though.
Don't use Windows, use OS's designed with security in mind.
I'd agree with this - certainly for mission-critical systems anyway. However, *all* OSes must be kept patched and up to date - a 4 year old Linux distribution is probably just as vulnerable as a 4 year old Windows release, it's only when you keep them patched up to date that Linux gets significantly more secure than Windows.
For workstations, Windows is sometimes a necessary evil but I think in most cases you *can* ditch Windows in favor of a better OS (Linux or consider OS X if Linux won't run the software you need).
Use SELinux or equivalent on mission critical nodes.
SELinux is still far from perfect on current distributions - certainly under Fedora Core 3 the supplied policies are too restrictive in a number of cases (Apache can't do a lot of stuff you want it to do, etc.). Whilest you _could_ rewrite the SELinux policies, you probably need a brain the size of a planet.
And secondly, educate the users and gain a culture of safety.
This is probably _the_ most important point. No matter how much you secure the software, the users are always a weak point. For the servers this isn't a big deal coz anyone who can log into them has (hopefully) got a clue. But you don't need to compromise the servers to cause disruption - once a single workstation has been compromised (maybe the user wanted to look at the cool new screensaver someone mailed them, whcih turned out to be a trojan) then your network is unsafe - your firewall won't do you much good now.
Your value for Boltzmann's constant is way too high.
Hmm, ok, I got the figure from Wikipedia which recons you multiply the energy in eV by 11,605 to get the temperature in K. Is that wrong or am I misunderstanding it somehow?
Also, temperature isn't necesarily well-defined or meaningful at these very low densities (that's right, density, screw pressure). Heat content might be a better way to look at things.
Well, that was sort of my point that temperature is probably not such a useful thing to measure at such low densities.
I don't think 'hot' is used to describe a location that is 7 billion miles from the sun
:).
Actually, "hot" (or temperature) is really describing the energy of the particles in the area. Inside the solar system, the solar wind is moving at pretty high speeds - wikipedia suggests energies of 500 KeV. Using the Boltzmann Constant we get 500,000 x 11,605 = 5.8 billion degrees K (Sounds a lot - can some astrophysacists check my figures please
Once you get to the termination shock, the solar wind is moving at much slower sub-sonic speeds. Not sure what energies we're talking about here but they're going to be a *lot* lower... A bit of googling suggests He energies somewhere around the 5.2 KeV area (5,200 x 11,605 = 60 million degrees K).
Of course, although the matter may be "hot", there isn't much of it - the low density of matter means that there isn't much "heat" (compare - a cigarette is "hot" (it's gonna burn you) whereas a central heating radiator is not as hot but generates more "heat" (it'll warm your room better than the cigarette because it's total energy output is much greater, even though it's temperature is less)).
Disclaimer: IANAAP (Astro-Physacist) so the above could be crap, but that is how I understand it.
Of course, admins love Windows. They hate Linux and Macs, because companies running these systems don't need as many admins.
I was installing some (Linux based) networking kit at a reasonably big customer a few weeks ago.
Now, from my perspective, if someone told me that one of my Linux systems had been compromised I'd say "oh shit" and be dropping everything to take it offline and reinstall, or at least clean it ASAP. (Yep, compromises _do_ happen on Linux kit but are reasonably infrequent if you keep stuff up to date, and 9 times out of 10 are caused my human error such as forgetting to turn on scheduled yum updates or leaving a weak password on a server).
While I was doing this install, I had to inform the customer that it looked like one of his (many) windows servers had been hit by some worm (lots of network traffic hitting random addresses on the internet on obviously microsofty ports). The reaction kind of shocked me - it was pretty much "oh well, happens all the time, I'll fix it later".
I think this says a lot about Windows admins - they have so many security problems that a random exploit (which, for all they know, could be posting confidential data all over the place) is considered pretty much the norm and unimportant.
With a helicopter, for instance, as long as you "float" over the deck without hitting anything, you can land anywhere.
I'm guessing that if the deck is going up and down by 20 feet every few seconds then "floating" over the deck without hitting it gets somewhat tricky. I know nothing of actually landing choppers in such conditions, but I understand that being lowered from a chopper onto the deck can be very dangerous, leading to broken bones or even death (large waves really do lift ships reasonable distances at quite a speed).
Remember when you started to pay cable TV to get rid of the commercials?? remember how did you feel when your cable company started to put commercials in their channels?
Errm, I dunno about you, but I don't pay Google...
I think we need to face the fact that Google now exists primarily to sell ads.
Is this not how it's always been? The primary reason for a company to exist is to make money. Google makes all it's money through ads, so obviously Google's primary objective is to sell ad space. Ad space is more valuable if more people see it, so if google continue to provide an excellent service then they will attract more people and make more money from the ads - everyone wins.
Right now, there are a zillion anonymous proxies on wired connections.
It's actually pretty hard to _guarantee_ you are anonymous on the internet. If you use an anonymous proxy then your IP will be hidden from the end web server you are contacting, but there is *no way* to know if the anonymous proxy is keeping logs. The authorities can track your web accesses back to the proxy, and if the proxy is keeping logs then it's very easy for the authorities to get the logs through a court order and tie the web server accesses to your real IP address.
The only way to guarantee an anonymiser isn't holding incriminating evidence against you is if you administer it yourself... and in that case you're probably screwed anyway.
I don't understand this idea that poor UI design is "OK for techies." We all benefit from good design, even power users.
Yes, techies benefit from good UI design too. However, techies can get by with an over-complex and inconsistent UI whereas non-techies would probably just get confused, frustrated and give up.
To most techies, having a feature-rich application is often more important than having a really simple UI, whereas to most non-techies they are happy with not so many features and a really simple UI.
Of course, "feature-rich" and "really simple" don't always (often?) go together very well. I think the current trend in Gnome is a perfect example of this - they're ripping out all the useful functionality because they claim it complicates things too much. (I used to use Gnome exclusively... then it lost all the features I liked so I stopped).
The thing that frustrates me more than anything else is that there's no excuse for this, except for the choices made by the project leader.
:)
I don't think this is down to a choice my Isaac but simply down to the ad-hoc nature with which FOSS projects such as Myth are developed. Different bits come from different people and so consistency can be limited. What they actually need is a "UI guru" or a organised team of UI people to sanitise the UI and new elements as they are implemented, however that involves interested parties to have time to do that sort of thing and many (techies) perceive the UI as somewhat secondary to the "cool stuff" (I am often no different with projects I'm working on, preferring to play with cool new stuff instead of boring sanitisation of other people's work).
There's a lot to be read into the fact that users will now be able to vote on which features should be prioritized for development.
I must make a very important point here that many people (users) just don't get: with FOSS software, everyone developing it is freely putting their time into the project and getting nothing in return except the features they are implementing. This means that in general, developers working on a FOSS project will only implement features that they themselves want. So basically, if you want something doing then you may well have to do it yourself, you certainly shouldn't expect someone else to give up their time to do it.
Commercial projects, OTOH, are not bound by such problems and so a developer may well develop a feature you want. However, this only works to a point - larger companies (i.e. those who don't significantly benefit from a minority group buying a product) are unlikely to implement features for individual users and in many cases get too much input from the marketting people who want "cool" stuff people are never going to use or want to rip out functionality on the grounds that it "overcomplicates" it. When was the last time Microsoft implemented a feature you asked for?
I think there may well be an advantage in FOSS projects which have commercial backing so long as the commercial side doesn't have _too much_ influence over the featureset, etc (Asterisk seems to be a good example of an FOSS project with commercial backing)
Oh yes, I know I shouldn't reply to myself but a point I missed:
:(
There are no CAMs available for decoding Sky channels, so you have to use a normal Sky box to decode to analogue and then reencode to MPEG4 instead of just using a DVB-S card to suck the MPEG2 data straight off the satellite dish. This sucks but I don't think Ofcom (or whoever) is likely to force Sky to sell a CAM, which gives Sky+ a bit of an advantage.
Gotta disagree. Myth is nice but is still FAR lacking in many ways - UI and ease of development in particular (speaking from some experience).
I find Myth pretty feature-rich, and it certainly seems to screw up and forget to record stuff less frequently than the Sky+ boxes some of my friends have.
You might be right about the UI to some extant - it's mostly ok for the techie but probably not so suitable for the general public (but then are the general public going to build their own Myth box or just buy one of the commercial PVRs?).
I think the main problem with the UI from my point of view is the recording priorities stuff - I don't like having to juggle integer priorities for all my programs and would prefer to just see a list of shows ordered by priority and be able to move a show up and down the list.
There is also some inconsistency with key bindings too - most of the UI looks in the key bindings database to find out which key is "select", whcih is "play", etc. However, some parts of the UI make assumptions instead - i.e. expecting Enter to be "select". But that's reasonably minor and probably doesn't affect most people.
I've not really done any UI development for Myth (just added a few controls to some of the setup screens...), although I did write some of the back end code (A/V synchronisation routines, etc) and can't say it was that hard to implement, despite not really being a C++ coder - I usually just use C so there was a slight learning curve there.
I like Myth, but it has many warts, and missing program guide data is not one of them.
I use the RadioTimes listings and I have to say that everything has got a *lot* better since RT started providing machine readable listings - the site scraper used to take hours and every so often they'd change something that broke it. There is still the occasional problem that programmes which are rerun several times during the 2 week period you get listings for sometimes don't have matching descriptions or subtitles so you get 2 recordings but for the most part it's not bad. Of course I'd like radio listings too (used to get them from the scraper but they don't provide machine readable radio listings).
whatever the material is in old smoke detectors
The isotope you are looking for is Americium.