Especially when people try to pitch it against apt-get.
My experience of apt-get has been that it has pretty terrible ways of dealing with broken dependencies in the RPM database. When I tried it on a database with broken deps it decided to automatically "fix" the database by removing any packages that had any broken ancestoral dependencies. The result was that it automatically uninstalled the entire distribution - I've never touched it again since this experience. Whilest I have had problems with yum, it's never tried to do anything quite so crazy.
I think all the SIP phones I've used (both softphone and hardware) have come with STUN pre-configured so they should Just Work. My only real complaint with this is that the STUN support on my UTStarcom F1000G is a bit broken and it sends invalid SIP packets if you're using a STUN server and you're *not* going through a NAT (a bug that UTStarcom won't acknowledge no matter how many dumps of the broken SIP traffic I send them).
The point where STUN starts to become a real issue is when you want to call both phones on the internet (e.g. going through your NAT) and phones within your network (not going through a NAT). In this case the only real solution is to use a SIP proxy at your network edge. How does Skype handle this - does your traffic end up going out the the internet even when calling someone within your LAN?
"How do I call your phone?"
I'm not sure what's hard about this - to call a SIP phone you just enter the SIP address (e.g. sip:foo@example.com - you can even click sip hyperlinks in web pages, etc.). And once you've entered your account details for whatever PSTN breakout service provider you've chosen you just dial a PSTN number.
SIP is not workable in a modern NAT environment. I hate to break it to you, but the average user doesn't want to deal with the well-documented firewall woes it brings.
Most SIP clients support STUN (many are configured to use STUN out of the box). Whilest you can't work though *all* NATs with this method, the number of people who are using symmetric NAT (which you can't use SIP through) is very very small.
Note that Asterisk implemented IAX2 specifically because of SIP's crappiness.
IAX2 was developed to solve some problems that SIP was never designed to solve and most people don't need to care about (i.e. interworking between PBXes which share dialplans, etc). Admittedly the "shove everything over one port" design is nice, but it's not a standard protocol which makes it rather non-interoperable (similar to Skype) when you want to phone arbitrary contacts.
And Skype just works. No SIP softphone that I've ever seen just sets up and works like it.
I've had no problems with Ekiga and SJPhone - just bung in my SIP registration server address and login details. Pretty similar to setting up an email client.
Although most of the time I use an 802.11 SIP phone so that I don't have to be near a computer.
I've used GnomeMeeting extensively in the past, from before it was SIP-based, until they became Ekiga. I guess this is new. It's still not at all obvious on their web page.
Ekiga is a standard SIP phone, so works with any SIP PSTN breakout provider (of which there are hundreds). I'm sure they don't feel the need to document it any more than the Thunderbird developers feel the need to mention that Thunderbird can be used to send email through a service provider of your choice.
Yeah, ok, so you have to make a decision as to which service provider you use - I guess if you just want to be forced into using a specific service provider then SIP isn't for you. On the other hand, I prefer not to be locked into a single vendor - competition is good for a market, it improves quality of service and drives down costs.
Also, does it have the equivalent of SkypeIn?
Again, there are hundreds of PSTN to SIP gateways - go pick one that suits your needs. VoipUser, SIPGate, BT, etc to name a few.
Really, it seems that your main complaint about SIP is that because it's a standard protocol you get a choice and you would prefer someone else made the choice for you. How do you deal with the PSTN world where you can go and buy a phone and *shock* use it with any service provider? You you require your POTS phone to also tie you in to a specific service provider?
The real way to solve NAT issues is through centralization or upnp.
I'd counter that the real way to solve NAT issues is to throw NAT away and use IPv6. NAT is a horrible kludge that's increasingly causing more and more problems as p2p apps become more popular - the sooner people realise this the better.
Not sure what's hard about SIP - put in your authentication details and registration server address and that's about it - not really any harder than configuring an email client. The only real comment I've heard is that people have to make a choice about what PSTN breakout provider to use (oh, the hardship of shopping around for a good deal instead of just being forced to use a single service provider).
2. No NAT issues (SIP is retarded with NAT - check out how SDP works).
I don't think SDP is "retarded" - there's just no good way of doing peer-to-peer through NAT reliably since NAT is specifically designed to break the end-to-end architecture of the Internet and turn it into a client/server network.
Skype works around the NAT problems by turning other people's Skype clients into servers to route the traffic (i.e. you're nolonger peer-to-peer) - something that I see as a pretty bad idea since you're then dependent on the quality of an arbitrary third party's network connection. Far better to fix your own NAT so it doesn't break your peer-to-peer traffic.
4. Same client, multiple platforms thanks to Qt.
There are plenty of multi-platform SIP clients available. SJPhone and X-Lite spring immediately to mind.
5. Voice quality is related to codec, not call setup protocol, which is what SIP is, so your voice quality comment is senseless.
Voice quality is also dependent on the SIP endpoint - having too small a jitter buffer leads to crumby audio in hostile network conditions. Sensible SIP endpoints have dynamic jitter buffers which size themselves according to network conditions, but some software/providers are poor and use fixed, small, jitter buffers. From reports I've heard, Skype seem to use very large jitter buffers which leads to good audio quality at the expensive of a massive latency.
Seamless integration with landlines.
I'm not sure how Skype is more seamless than SIP when dealing with the PSTN - could you clarify?
7. Lots of features (video, chat, etc., all encrypted).
Funny - I could've sworn I saw video, chat, etc in Ekiga (my SIP client)...
SIP is mostly a joke.
If SIP is a joke, please explain why all the western telecoms companies are investing billions in SIP as a replacement for the aging SS7/TDM networks.
I'm inclined to say that Skype is a joke since it relies on a closed protocol, meaning that it doesn't interoperate with any other VoIP application/service provider and locks the customer into a single vendor. The only thing Skype have got right is the marketing... The Internet was built on open standards, using a closed standard for telecommunications is just crazy.
I'm afraid you are spreading mis-information. SIP is a peer to peer protocol.
With SIP, when you call someone@example.com, your user agent looks up the DNS records for example.com and sends SIP traffic to the associated server asking to start a call. The software at example.com can either be a phone (i.e. you're doing real P2P at this stage) or more frequently a server (which is essentially acting as a router. Skype does the same thing, otherwise you'd need to know stuff like the IP address of the person you're phoning).
If example.com is a server, it will send SIP traffic to the actual phone to start the call (or another server - in the eyes of the SIP protocol there is no difference). Usually the 2 end points will then send RTP traffic (which carries the voice, video, etc) directly to each other. I.e. the actual voice traffic is going peer to peer. Of course this doesn't _have_ to happen - you can set up the servers in the middle to request the RTP traffic be sent via them.
So to summarise, the *usual* setup involves routing the signalling via one or more servers (exactly like Skype) and sending the voice traffic directly between the peers. But SIP doesn't restrict you to this architecture, and you can quite happily make completely peer-to-peer calls with no 3rd party servers involved so long as you know the address of the phone you're calling.
By contrast, Skype is often _not_ peer to peer since it hijacks other users' internet connections to route voice traffic in some cases.
Also, the skype client is supposed to throttle the amount of bandwidth it uses
You will usually find the Skype client uses a lot more bandwidth than a SIP client since it is often used to relay unrelated traffic for other people's calls.
ACPI is one of those things that's generally considered to be a completely screwed design to begin with. And then it doesn't help that most of the BIOS writers completely fail to comply with the spec (although this is getting better now).
and of all the Linux distros I tried on it, only Kubuntu came close to doing it right.
I have Fedora Core 3 on a P166 laptop (which is used as a simple X terminal) and that goes to sleep and wakes up with no problems at all (although it is APM, not ACPI). Press the "sleep" button and 5 seconds later it's asleep. Press the power button and 5 seconds later it's awake again (a further ~5 seconds to re-associate with my 802.11 accesspoint and get a DHCP lease).
My MythTV box also hibernates successfully, although the crappy Gigabyte motherboard takes forever to POST on wakeup (over 15 seconds to POST before it starts reloading the memory from disk!). Also, Wake On LAN utterly fails to work on that board. I won't be buying Gigabyte motherboards again - all 3 of the Gigabyte boards I've bought in the past 5 years have crap ACPI, with things like WoL not working, no temperatures being reported, etc. The reason my MythTV system suspends to disk instead of RAM is because the ACPI BIOS completely powers off the whole machine (RAM and all) when asked to go to sleep.
I suspect Apple are so successful with power management because they have a lot more control over the hardware - they don't have to kludge up workarounds to deal with broken ACPI - if the firmware is broken they have the ability to fix it.
What, you think that any ISP would actually allow downloads fast enough to use over 100baseT?
Believe it or not, some people use LANs for things other than accessing the internet... The internet connection speed becomes unimportant if the network is actually a SAN.
Really, even full 10baseT (as an obtainable download speed, not just the home->CO link speed) would be an improvement to many people.
We're reaching the point now where I've stopped caring so much about download speed (I have an 8Mbps DSL) - upload speed is becoming a serious headache since on most ADSL lines (at least in the UK) it tops out at ~340Kbps. At that upload speed you're talking about ~45ms per MTU sized (1500 byte) packet - that's quite a lot of latency jitter and can cause serious problems for realtime applications such as VoIP, which often have jitter buffers of only around 100ms long.
greeted with a large banner at the top and an audio recording saying my computer was "infected" with internet explorer
The example you cite is bad because of the wording. However, I have received abusive emails from people surfing my website in IE before now. The reason: I put a warning message on my web pages when viewed in IE recommending that people upgrade to FireFox since it renders web pages (more) correctly and is more secure. And indeed, my website won't display correctly in IE - I don't want to put the time in to work around IE's brokenness just so people can continue to use a truely broken browser. Nor do I own a Windows machine to test IE on.
When MS decide to make IE standards complient to a similar level to FireFox and Opera, *and* produce a version that I can run for free, I'll start testing my site in IE and working around the minor bugs that are left. My website isn't commercial, I don't make any real money off it, so if someone wants access to the information on the site they should show the courtesy of using a sane browser rather than bitching to the site owner.
Your unsubstantiated claim that cash is best for low-value transactions is absurd!
Can you really use a credit card for all transactions in the US? In the UK most places won't accept cards for transactions under 5ukp since the credit card company charges the retailer for every transaction. Also, what to you do if you're giving a small amount of money to another individual? Do your friends all carry around credit card readers?
if I were an advertiser I'd find it a tough sell to pay any extra for an uncertain market.
It may not be worth the advertisers paying extra *now*, but in a few years time, when a large proportion of people have HDTVs and most broadcasters are showing HDTV, the broadcasters who are sticking to SDTV because "there's no commercial reason to do HDTV" are going to lose out.
Think about it, $broadcaster[0] is showing $sporting_event in HD, $broadcaster[1] is showing $sporting_event in SD - which one are you going to watch, assuming you have an HDTV (all other things being equal)? Although consumers/advertisers might not want to pay extra for HDTV, they are going to choose it over SDTV if there's no price difference - this means that in the long run, any broadcaster who doesn't embrace the new technology is going to lose out.
This is of course because spreading spam costs too little to be worried about pre-selecting the audience.
Whilest spam is by far the worst case, all direct marketting suffers from this problem to some extent. Very little of the crap that's shoved through my door, SMSed or telemarketted to me is actually relevent to me.
At least in the UK we have some of the direct marketting a little more under control (unsolicited SMS messages are illegal... although some do still get sent. Telemarketting to phones registered with the telephone preference service is illegal, not that this seems to stop some telemarketters).
Of course, if it were down to me, direct marketting of all forms would be completely illegal - it's of no benefit to the consumer, unlike things like TV advertising which benefit the consumer by paying for the TV channel.
So really, whilest increased cost would certainly reduce the problem to some extent, the other direct marketting methods show that it will by no means eliminate the problem of untargetted advertising.
Even if it does happen, though, it will only go to show that captchas aren't the way to get rid of spam, bots etc.
I would certainly like to see the end of captchas, and I have resisted using them on my own sites. They are really bad for accessibility and therefore illegal in many situations and just generally unfair to anyone who can't solve captchas (whether that be by disability or browser choice). However, I have yet to see any other technology able to do the job.
So you're suggesting that every product which I don't want and I don't buy, I am boycotting? I've been boycotting prunes all my life and didnt realise it!
No - if you _want_ a product but choose not to buy it because you disagree with something you are boycotting it.
i.e. you want that music track but avoid buying it because it is DRM'd.
When you buy that CDROM drive, you have paid a licencing fee. Without paying that licencing fee, you can't listen to your CD.
Yes. And with that CDROM drive I can do what the hell I like with it. For example - I can install it in my own music playing device, or I could disassemble it or modify it - whatever.
With a DRM system that is usually not an option - the licence will restrict what I can do with the decoder, and the EUCD makes it illegal for me to reverse engineer it.
choosing not to buy a product is not really "boycotting"
I suggest you look up the meaning of the word. boycott To abstain,either as an individual or group, from using, buying, or dealing with someone or some organisation as an expression of protest or as a means of coercion.
Because failing to pay for it whilest still owning it would be illegal.
(and for the record, my entire music library is on CD and ripped to Vorbis so it's not going to suddenly stop working. But if I had purchased it from an online store doing DRM'd music then it's vary possible it could all just stop working one day)
What you propose just results is massive duplication of research.
So basically we have 2 options: 1. Disallow software patents, people wishing to protect their work will keep it secret and other people can come up with their own (probably very similar) solutions to the same problem. 2. Allow software patents and therefore prevent *anyone* else from producing anything similar on their own.
Option (1) seems like the better option to me since at least it allows third parties to come up with a solution. Remember that in most cases the patent holder either won't licence their technology at all, will licence it for infeasable amounts of money, or put very restrictive terms on the licence (what good is the ability to use some technology if you're not allowed to integrate it into your FOSS project?)
And this is assuming you even realise you're infringing someone's patent. Remember, you're still infringing a patent even if you came up with the idea on your own - all too often a product is developed independently, becomes very successful and then the producer is sued for infringing a patent that they had never heard of. It's now got to the point where it's pretty much impossible to write a piece of software without infringing _someone's_ patent.
Also, from my experience the threat of people suing for patent infringement often motivates corporations to keep source code closed which they would otherwise be happy to open to the public - this is certainly not a good thing.
Patents have been turned into ammunition for large corporations - having 100,000 patents prevents the guy who only has 80,000 patents from suing you. Unfortunately the little guy who's got no patents and no money for lawyers gets completely squashed in the process. Maybe patents are sometimes good for small inventors, but they are open to abuse by large organisations. And even if you are in the right and have prior art, how the hell are you, as a single person on your own, going to be able to defend yourself in court against some huge corporation such as IBM, Microsoft, HP, etc?
I've not tried it, but apparently you can use a Dragon CAM module
You can, it's illegal, very expensive and may stop working at any moment if Sky decide to update the algorithm again.
I would like to see Ofcom order Sky to at least offer a reasonably priced CAM, or better - publish the decryption algorithm so that it could be implemented in software. Publishing the algorithm shouldn't risk their security since it would still require the smart card. This is certainly something they could require Sky to do under anti-monopoly laws, especially since both Channel 4 and Channel 5 (which are traditionally free) currently use VideoGuard (although Channel 4 is expected to move to the Astra 2D transponder and go free to air some time in 2007/8) and Ofcom are supposed to be promoting the move to digital.
Having to pay Sky to receive Sky channels is fine, but having to pay Sky to receive random other channels that are nothing to do with Sky is pretty poor.
If you buy a DRM-free CD, a licence fee is paid to the inventors of CDs.
That's somewhat different - a CD is useful to *me*, DRM is useful only to the content provider.
Besides, I was really talking about the licence required for the decoder, not the licence that the publisher has to pay to encrypt the content. I don't care what technology the publisher uses so long as I can play the content for free without unreasonable restrictions on my legal rights. At the moment this is not the case - in order to play DRM'd content I must have a licence to use some kind of decoder, even though I've already paid for the right to use the content itself. This is very different from a CD - I can go to the shops and buy a fully legit CDROM drive, which I can do with what I want - I can install it in any system I want and I can disassemble it and modify it - show me a DRM system which I can do the same with.
If you buy a car, a number of licence fees are paid to inventors of technology within that vehicle.
When I buy a car I expect it to work without further cost - I don't expect to buy the car and then have to pay a further licence fee in order to use it. If I am to play DRM'd content I must not only buy the content itself but I must have a licence for a player in order to use the content.
The point is you don't have to play their content. You can choose not to care about that product.
Boycotting a product only works if there is an alternative you can buy instead. When *all* music is DRM'd you have to choose between accepting the DRM or doing without *any* music. Guess which the masses will be doing? The little guy who wants to stand up for his rights loses out because the masses just accept it and noone will protect the rights of those who won't accept it. Part of the deal with DRM is that some of the more serious effects won't be noticed by the masses for many years, by which time it'll be too late to reverse the trend anyway.
There are also laws that prevent anti-competetive practices - why are these not being applied to prevent this situation?
And for the record - I don't buy DRM'd content. If something isn't going to work on an open source player, it's completely useless to me.
But of course, any system which checks the validity of DRM licenses would be attacked as an invasion of "privacy."
If you have to validate the licence by contacting a remote server, not only is it an invasion of privacy, but if the remote server is inaccessible you lose the ability to use the content that you have licenced. So if the licensor goes out of business, you have ultimately lost access to everything you have paid for. I've brought up this comment with regards to DRM schemes like iTMS before, and invariably I get a comment like "There's no way a big company like Apple would go under" - what a naive claim. Big companies go under all the time, you only have to look at Enron for a recent example.
For me, this is the big deal - if my entire music library that I have *paid* for suddenly stops working, I'm going to be pretty pissed off.
Also, the ability to use DRM'd content *now* is a big deal. If I have paid for some content, why must I also be required to pay a licence fee to the owner of the DRM technology? This is usually going to either tie me to specific hardware (e.g. I'd have to buy a commercial BluRay player and HDCP capable TV), cost me an infeasable amount of money (wanna try asking Microsoft for a licence to decode WMP's DRM in your personal project?) or tie me to a specific operating system (why should I be required to buy Windows - an operating system that is completely useless to me - and a new computer to run it on, just so I can play some Microsoft DRM protected content? Seems rather anticompetetive to me - what we effectively have is a cartel of corporations who are doing their level best to lock anyone else out of the market.
Here is a real world example: I use MythTV as my PVR with a DVB-S card. I cannot use this system to pick up much of the satellite programming here in the UK because it is broadcast using NDS-Videoguard encryption. The *only* way you can use such broadcasts with a PVR (without going through the analogue hole) is to buy Sky's own Sky+ PVR system. Sounds anticompetetive doesn't it? Sky have a monopoly on satellite enabled PVRs in the UK because noone else can legally produce a PVR system that can receive many of the satellite channels. This doesn't just apply to Sky's own channels either - many channels that are touted as "free" are still encrypted using this system and you still have to buy Sky licenced equipment to receive these channels. (And before anyone suggests that Sky own the satellite, they don't - SES own the Astra 2 constellation.)
From BBC : "The average citizen in the UK is caught on CCTV cameras 300 times a day."
I've heard this statistic before. Whilest it may be accurate for those people who walk through a city centre every day (especially people who commute on the Tube in London), It seems a rediculously high figure for the rest of us.
About the only time I'm going to get caught on a security camera during most days is when driving to work - my trip to work involves about 10 miles on the motorway and 15 miles on an A road. There are no cameras on the A road (it's in the middle of nowhere). Of course, when I go shopping there will be more cameras, but it seems unlikley this ever gets anywhere near 300.
It seems more likley that this statistic is really "the average person working in London".
Slashdot Mirror
Especially when people try to pitch it against apt-get.
My experience of apt-get has been that it has pretty terrible ways of dealing with broken dependencies in the RPM database. When I tried it on a database with broken deps it decided to automatically "fix" the database by removing any packages that had any broken ancestoral dependencies. The result was that it automatically uninstalled the entire distribution - I've never touched it again since this experience. Whilest I have had problems with yum, it's never tried to do anything quite so crazy.
"What's STUN?" she would say.
I think all the SIP phones I've used (both softphone and hardware) have come with STUN pre-configured so they should Just Work. My only real complaint with this is that the STUN support on my UTStarcom F1000G is a bit broken and it sends invalid SIP packets if you're using a STUN server and you're *not* going through a NAT (a bug that UTStarcom won't acknowledge no matter how many dumps of the broken SIP traffic I send them).
The point where STUN starts to become a real issue is when you want to call both phones on the internet (e.g. going through your NAT) and phones within your network (not going through a NAT). In this case the only real solution is to use a SIP proxy at your network edge. How does Skype handle this - does your traffic end up going out the the internet even when calling someone within your LAN?
"How do I call your phone?"
I'm not sure what's hard about this - to call a SIP phone you just enter the SIP address (e.g. sip:foo@example.com - you can even click sip hyperlinks in web pages, etc.). And once you've entered your account details for whatever PSTN breakout service provider you've chosen you just dial a PSTN number.
SIP is not workable in a modern NAT environment. I hate to break it to you, but the average user doesn't want to deal with the well-documented firewall woes it brings.
Most SIP clients support STUN (many are configured to use STUN out of the box). Whilest you can't work though *all* NATs with this method, the number of people who are using symmetric NAT (which you can't use SIP through) is very very small.
Note that Asterisk implemented IAX2 specifically because of SIP's crappiness.
IAX2 was developed to solve some problems that SIP was never designed to solve and most people don't need to care about (i.e. interworking between PBXes which share dialplans, etc). Admittedly the "shove everything over one port" design is nice, but it's not a standard protocol which makes it rather non-interoperable (similar to Skype) when you want to phone arbitrary contacts.
And Skype just works. No SIP softphone that I've ever seen just sets up and works like it.
I've had no problems with Ekiga and SJPhone - just bung in my SIP registration server address and login details. Pretty similar to setting up an email client.
Although most of the time I use an 802.11 SIP phone so that I don't have to be near a computer.
I've used GnomeMeeting extensively in the past, from before it was SIP-based, until they became Ekiga. I guess this is new. It's still not at all obvious on their web page.
Ekiga is a standard SIP phone, so works with any SIP PSTN breakout provider (of which there are hundreds). I'm sure they don't feel the need to document it any more than the Thunderbird developers feel the need to mention that Thunderbird can be used to send email through a service provider of your choice.
Yeah, ok, so you have to make a decision as to which service provider you use - I guess if you just want to be forced into using a specific service provider then SIP isn't for you. On the other hand, I prefer not to be locked into a single vendor - competition is good for a market, it improves quality of service and drives down costs.
Also, does it have the equivalent of SkypeIn?
Again, there are hundreds of PSTN to SIP gateways - go pick one that suits your needs. VoipUser, SIPGate, BT, etc to name a few.
Really, it seems that your main complaint about SIP is that because it's a standard protocol you get a choice and you would prefer someone else made the choice for you. How do you deal with the PSTN world where you can go and buy a phone and *shock* use it with any service provider? You you require your POTS phone to also tie you in to a specific service provider?
The real way to solve NAT issues is through centralization or upnp.
I'd counter that the real way to solve NAT issues is to throw NAT away and use IPv6. NAT is a horrible kludge that's increasingly causing more and more problems as p2p apps become more popular - the sooner people realise this the better.
1. Dead easy setup.
Not sure what's hard about SIP - put in your authentication details and registration server address and that's about it - not really any harder than configuring an email client. The only real comment I've heard is that people have to make a choice about what PSTN breakout provider to use (oh, the hardship of shopping around for a good deal instead of just being forced to use a single service provider).
2. No NAT issues (SIP is retarded with NAT - check out how SDP works).
I don't think SDP is "retarded" - there's just no good way of doing peer-to-peer through NAT reliably since NAT is specifically designed to break the end-to-end architecture of the Internet and turn it into a client/server network.
Skype works around the NAT problems by turning other people's Skype clients into servers to route the traffic (i.e. you're nolonger peer-to-peer) - something that I see as a pretty bad idea since you're then dependent on the quality of an arbitrary third party's network connection. Far better to fix your own NAT so it doesn't break your peer-to-peer traffic.
4. Same client, multiple platforms thanks to Qt.
There are plenty of multi-platform SIP clients available. SJPhone and X-Lite spring immediately to mind.
5. Voice quality is related to codec, not call setup protocol, which is what SIP is, so your voice quality comment is senseless.
Voice quality is also dependent on the SIP endpoint - having too small a jitter buffer leads to crumby audio in hostile network conditions. Sensible SIP endpoints have dynamic jitter buffers which size themselves according to network conditions, but some software/providers are poor and use fixed, small, jitter buffers. From reports I've heard, Skype seem to use very large jitter buffers which leads to good audio quality at the expensive of a massive latency.
Seamless integration with landlines.
I'm not sure how Skype is more seamless than SIP when dealing with the PSTN - could you clarify?
7. Lots of features (video, chat, etc., all encrypted).
Funny - I could've sworn I saw video, chat, etc in Ekiga (my SIP client)...
SIP is mostly a joke.
If SIP is a joke, please explain why all the western telecoms companies are investing billions in SIP as a replacement for the aging SS7/TDM networks.
I'm inclined to say that Skype is a joke since it relies on a closed protocol, meaning that it doesn't interoperate with any other VoIP application/service provider and locks the customer into a single vendor. The only thing Skype have got right is the marketing... The Internet was built on open standards, using a closed standard for telecommunications is just crazy.
Skype is P2P (unlike SIP)
I'm afraid you are spreading mis-information. SIP is a peer to peer protocol.
With SIP, when you call someone@example.com, your user agent looks up the DNS records for example.com and sends SIP traffic to the associated server asking to start a call. The software at example.com can either be a phone (i.e. you're doing real P2P at this stage) or more frequently a server (which is essentially acting as a router. Skype does the same thing, otherwise you'd need to know stuff like the IP address of the person you're phoning).
If example.com is a server, it will send SIP traffic to the actual phone to start the call (or another server - in the eyes of the SIP protocol there is no difference). Usually the 2 end points will then send RTP traffic (which carries the voice, video, etc) directly to each other. I.e. the actual voice traffic is going peer to peer. Of course this doesn't _have_ to happen - you can set up the servers in the middle to request the RTP traffic be sent via them.
So to summarise, the *usual* setup involves routing the signalling via one or more servers (exactly like Skype) and sending the voice traffic directly between the peers. But SIP doesn't restrict you to this architecture, and you can quite happily make completely peer-to-peer calls with no 3rd party servers involved so long as you know the address of the phone you're calling.
By contrast, Skype is often _not_ peer to peer since it hijacks other users' internet connections to route voice traffic in some cases.
Also, the skype client is supposed to throttle the amount of bandwidth it uses
You will usually find the Skype client uses a lot more bandwidth than a SIP client since it is often used to relay unrelated traffic for other people's calls.
My stupid Latitude has such a buttfargled ACPI
ACPI is one of those things that's generally considered to be a completely screwed design to begin with. And then it doesn't help that most of the BIOS writers completely fail to comply with the spec (although this is getting better now).
and of all the Linux distros I tried on it, only Kubuntu came close to doing it right.
I have Fedora Core 3 on a P166 laptop (which is used as a simple X terminal) and that goes to sleep and wakes up with no problems at all (although it is APM, not ACPI). Press the "sleep" button and 5 seconds later it's asleep. Press the power button and 5 seconds later it's awake again (a further ~5 seconds to re-associate with my 802.11 accesspoint and get a DHCP lease).
My MythTV box also hibernates successfully, although the crappy Gigabyte motherboard takes forever to POST on wakeup (over 15 seconds to POST before it starts reloading the memory from disk!). Also, Wake On LAN utterly fails to work on that board. I won't be buying Gigabyte motherboards again - all 3 of the Gigabyte boards I've bought in the past 5 years have crap ACPI, with things like WoL not working, no temperatures being reported, etc. The reason my MythTV system suspends to disk instead of RAM is because the ACPI BIOS completely powers off the whole machine (RAM and all) when asked to go to sleep.
I suspect Apple are so successful with power management because they have a lot more control over the hardware - they don't have to kludge up workarounds to deal with broken ACPI - if the firmware is broken they have the ability to fix it.
A huge percentage of legitimate email is random sentences with buzz words and a picture.
:)
I for one wouldn't mind emails containing managerspeak to be automagically binned
What, you think that any ISP would actually allow downloads fast enough to use over 100baseT?
Believe it or not, some people use LANs for things other than accessing the internet... The internet connection speed becomes unimportant if the network is actually a SAN.
Really, even full 10baseT (as an obtainable download speed, not just the home->CO link speed) would be an improvement to many people.
We're reaching the point now where I've stopped caring so much about download speed (I have an 8Mbps DSL) - upload speed is becoming a serious headache since on most ADSL lines (at least in the UK) it tops out at ~340Kbps. At that upload speed you're talking about ~45ms per MTU sized (1500 byte) packet - that's quite a lot of latency jitter and can cause serious problems for realtime applications such as VoIP, which often have jitter buffers of only around 100ms long.
Imagine a beowulf cluster of those....
:)
(sorry
greeted with a large banner at the top and an audio recording saying my computer was "infected" with internet explorer
The example you cite is bad because of the wording. However, I have received abusive emails from people surfing my website in IE before now. The reason: I put a warning message on my web pages when viewed in IE recommending that people upgrade to FireFox since it renders web pages (more) correctly and is more secure. And indeed, my website won't display correctly in IE - I don't want to put the time in to work around IE's brokenness just so people can continue to use a truely broken browser. Nor do I own a Windows machine to test IE on.
When MS decide to make IE standards complient to a similar level to FireFox and Opera, *and* produce a version that I can run for free, I'll start testing my site in IE and working around the minor bugs that are left. My website isn't commercial, I don't make any real money off it, so if someone wants access to the information on the site they should show the courtesy of using a sane browser rather than bitching to the site owner.
Give a man a fire and he'll be warm for a day.
Set a man on fire and he'll be warm for the rest of his life.
If I want to give them money, I write them a check
:)
People still use cheques? wow, how quaint
Your unsubstantiated claim that cash is best for low-value transactions is absurd!
Can you really use a credit card for all transactions in the US? In the UK most places won't accept cards for transactions under 5ukp since the credit card company charges the retailer for every transaction. Also, what to you do if you're giving a small amount of money to another individual? Do your friends all carry around credit card readers?
if I were an advertiser I'd find it a tough sell to pay any extra for an uncertain market.
It may not be worth the advertisers paying extra *now*, but in a few years time, when a large proportion of people have HDTVs and most broadcasters are showing HDTV, the broadcasters who are sticking to SDTV because "there's no commercial reason to do HDTV" are going to lose out.
Think about it, $broadcaster[0] is showing $sporting_event in HD, $broadcaster[1] is showing $sporting_event in SD - which one are you going to watch, assuming you have an HDTV (all other things being equal)? Although consumers/advertisers might not want to pay extra for HDTV, they are going to choose it over SDTV if there's no price difference - this means that in the long run, any broadcaster who doesn't embrace the new technology is going to lose out.
This is of course because spreading spam costs too little to be worried about pre-selecting the audience.
Whilest spam is by far the worst case, all direct marketting suffers from this problem to some extent. Very little of the crap that's shoved through my door, SMSed or telemarketted to me is actually relevent to me.
At least in the UK we have some of the direct marketting a little more under control (unsolicited SMS messages are illegal... although some do still get sent. Telemarketting to phones registered with the telephone preference service is illegal, not that this seems to stop some telemarketters).
Of course, if it were down to me, direct marketting of all forms would be completely illegal - it's of no benefit to the consumer, unlike things like TV advertising which benefit the consumer by paying for the TV channel.
So really, whilest increased cost would certainly reduce the problem to some extent, the other direct marketting methods show that it will by no means eliminate the problem of untargetted advertising.
Even if it does happen, though, it will only go to show that captchas aren't the way to get rid of spam, bots etc.
I would certainly like to see the end of captchas, and I have resisted using them on my own sites. They are really bad for accessibility and therefore illegal in many situations and just generally unfair to anyone who can't solve captchas (whether that be by disability or browser choice). However, I have yet to see any other technology able to do the job.
So you're suggesting that every product which I don't want and I don't buy, I am boycotting? I've been boycotting prunes all my life and didnt realise it!
No - if you _want_ a product but choose not to buy it because you disagree with something you are boycotting it.
i.e. you want that music track but avoid buying it because it is DRM'd.
When you buy that CDROM drive, you have paid a licencing fee. Without paying that licencing fee, you can't listen to your CD.
Yes. And with that CDROM drive I can do what the hell I like with it. For example - I can install it in my own music playing device, or I could disassemble it or modify it - whatever.
With a DRM system that is usually not an option - the licence will restrict what I can do with the decoder, and the EUCD makes it illegal for me to reverse engineer it.
choosing not to buy a product is not really "boycotting"
I suggest you look up the meaning of the word.
boycott To abstain,either as an individual or group, from using, buying, or dealing with someone or some organisation as an expression of protest or as a means of coercion.
So why did you pay for it?
Because failing to pay for it whilest still owning it would be illegal.
(and for the record, my entire music library is on CD and ripped to Vorbis so it's not going to suddenly stop working. But if I had purchased it from an online store doing DRM'd music then it's vary possible it could all just stop working one day)
What you propose just results is massive duplication of research.
So basically we have 2 options:
1. Disallow software patents, people wishing to protect their work will keep it secret and other people can come up with their own (probably very similar) solutions to the same problem.
2. Allow software patents and therefore prevent *anyone* else from producing anything similar on their own.
Option (1) seems like the better option to me since at least it allows third parties to come up with a solution. Remember that in most cases the patent holder either won't licence their technology at all, will licence it for infeasable amounts of money, or put very restrictive terms on the licence (what good is the ability to use some technology if you're not allowed to integrate it into your FOSS project?)
And this is assuming you even realise you're infringing someone's patent. Remember, you're still infringing a patent even if you came up with the idea on your own - all too often a product is developed independently, becomes very successful and then the producer is sued for infringing a patent that they had never heard of. It's now got to the point where it's pretty much impossible to write a piece of software without infringing _someone's_ patent.
Also, from my experience the threat of people suing for patent infringement often motivates corporations to keep source code closed which they would otherwise be happy to open to the public - this is certainly not a good thing.
Patents have been turned into ammunition for large corporations - having 100,000 patents prevents the guy who only has 80,000 patents from suing you. Unfortunately the little guy who's got no patents and no money for lawyers gets completely squashed in the process. Maybe patents are sometimes good for small inventors, but they are open to abuse by large organisations. And even if you are in the right and have prior art, how the hell are you, as a single person on your own, going to be able to defend yourself in court against some huge corporation such as IBM, Microsoft, HP, etc?
I've not tried it, but apparently you can use a Dragon CAM module
You can, it's illegal, very expensive and may stop working at any moment if Sky decide to update the algorithm again.
I would like to see Ofcom order Sky to at least offer a reasonably priced CAM, or better - publish the decryption algorithm so that it could be implemented in software. Publishing the algorithm shouldn't risk their security since it would still require the smart card. This is certainly something they could require Sky to do under anti-monopoly laws, especially since both Channel 4 and Channel 5 (which are traditionally free) currently use VideoGuard (although Channel 4 is expected to move to the Astra 2D transponder and go free to air some time in 2007/8) and Ofcom are supposed to be promoting the move to digital.
Having to pay Sky to receive Sky channels is fine, but having to pay Sky to receive random other channels that are nothing to do with Sky is pretty poor.
If you buy a DRM-free CD, a licence fee is paid to the inventors of CDs.
That's somewhat different - a CD is useful to *me*, DRM is useful only to the content provider.
Besides, I was really talking about the licence required for the decoder, not the licence that the publisher has to pay to encrypt the content. I don't care what technology the publisher uses so long as I can play the content for free without unreasonable restrictions on my legal rights. At the moment this is not the case - in order to play DRM'd content I must have a licence to use some kind of decoder, even though I've already paid for the right to use the content itself. This is very different from a CD - I can go to the shops and buy a fully legit CDROM drive, which I can do with what I want - I can install it in any system I want and I can disassemble it and modify it - show me a DRM system which I can do the same with.
If you buy a car, a number of licence fees are paid to inventors of technology within that vehicle.
When I buy a car I expect it to work without further cost - I don't expect to buy the car and then have to pay a further licence fee in order to use it. If I am to play DRM'd content I must not only buy the content itself but I must have a licence for a player in order to use the content.
The point is you don't have to play their content. You can choose not to care about that product.
Boycotting a product only works if there is an alternative you can buy instead. When *all* music is DRM'd you have to choose between accepting the DRM or doing without *any* music. Guess which the masses will be doing? The little guy who wants to stand up for his rights loses out because the masses just accept it and noone will protect the rights of those who won't accept it. Part of the deal with DRM is that some of the more serious effects won't be noticed by the masses for many years, by which time it'll be too late to reverse the trend anyway.
There are also laws that prevent anti-competetive practices - why are these not being applied to prevent this situation?
And for the record - I don't buy DRM'd content. If something isn't going to work on an open source player, it's completely useless to me.
But of course, any system which checks the validity of DRM licenses would be attacked as an invasion of "privacy."
If you have to validate the licence by contacting a remote server, not only is it an invasion of privacy, but if the remote server is inaccessible you lose the ability to use the content that you have licenced. So if the licensor goes out of business, you have ultimately lost access to everything you have paid for. I've brought up this comment with regards to DRM schemes like iTMS before, and invariably I get a comment like "There's no way a big company like Apple would go under" - what a naive claim. Big companies go under all the time, you only have to look at Enron for a recent example.
For me, this is the big deal - if my entire music library that I have *paid* for suddenly stops working, I'm going to be pretty pissed off.
Also, the ability to use DRM'd content *now* is a big deal. If I have paid for some content, why must I also be required to pay a licence fee to the owner of the DRM technology? This is usually going to either tie me to specific hardware (e.g. I'd have to buy a commercial BluRay player and HDCP capable TV), cost me an infeasable amount of money (wanna try asking Microsoft for a licence to decode WMP's DRM in your personal project?) or tie me to a specific operating system (why should I be required to buy Windows - an operating system that is completely useless to me - and a new computer to run it on, just so I can play some Microsoft DRM protected content? Seems rather anticompetetive to me - what we effectively have is a cartel of corporations who are doing their level best to lock anyone else out of the market.
Here is a real world example: I use MythTV as my PVR with a DVB-S card. I cannot use this system to pick up much of the satellite programming here in the UK because it is broadcast using NDS-Videoguard encryption. The *only* way you can use such broadcasts with a PVR (without going through the analogue hole) is to buy Sky's own Sky+ PVR system. Sounds anticompetetive doesn't it? Sky have a monopoly on satellite enabled PVRs in the UK because noone else can legally produce a PVR system that can receive many of the satellite channels. This doesn't just apply to Sky's own channels either - many channels that are touted as "free" are still encrypted using this system and you still have to buy Sky licenced equipment to receive these channels. (And before anyone suggests that Sky own the satellite, they don't - SES own the Astra 2 constellation.)
From BBC : "The average citizen in the UK is caught on CCTV cameras 300 times a day."
I've heard this statistic before. Whilest it may be accurate for those people who walk through a city centre every day (especially people who commute on the Tube in London), It seems a rediculously high figure for the rest of us.
About the only time I'm going to get caught on a security camera during most days is when driving to work - my trip to work involves about 10 miles on the motorway and 15 miles on an A road. There are no cameras on the A road (it's in the middle of nowhere). Of course, when I go shopping there will be more cameras, but it seems unlikley this ever gets anywhere near 300.
It seems more likley that this statistic is really "the average person working in London".