Yes, Nikon does add some obscurity by encrypting the WB setting. However, the original poster made it sound as if the entire file was being encrypted, which is a very different thing.
At best, encrypting the WB setting makes things mildly more annoying for if you're working with a 3rd party RAW converter. The reason I say this is because I know a number of photographers who don't use WB when shooting and instead handle that in post-production. Others may set the WB simply so that the JPEG preview looks reasonably close to what the final shot would look like.
That aside, let's look at what the WB setting actually does. It adjusts the colour based on the colour temperature of the lighting in order to remove any colour cast. A lot of people obsess over using custom WB settings but it's not really that big a deal. For outdoor shooting you probably want to maintain the character of the lighting so you don't want a custom WB. For instance, if you're shooting just around sunrise or sunset you want that nice yellow-orange light. Setting a custom WB will eliminate that look. Another example would be a message I saw on one of the photo lists I'm on, the poster was complaining that his snow shots didn't have the slight blue colour he wanted to make things look cold. The problem was he was using a custom white balance that was balancing the snow to be pure white instead of slightly blue. Simply using the "daylight" WB would have been the best option.
For indoor shooting you're most likely dealing with either tungsten or flourescent lighting. Both of these can usually be adjust for with standard WB settings.
So what's left? The main usefulness for custom WB settings is when you're working in mixed lighting. Say an indoor shot where you have tungsten and flourescent lighting as well as daylight coming in from windows. That's a tricky lighting situation. That's when you are going to want to set a custom WB. However, you can just as easily take a shot of your grey card (which you should be using to get an accurate WB setting in the first place). If you really want to get accurate colour you should shoot a frame of a colour chart (such as a Macbeth colour chart) and adjust off of that. In your RAW converter you can then use that shot to set your WB for the whole set of shots.
As for the DMCA, it sounds like it would set Nikon up to be in a position to sue if another company decrypts their WB information without permission but you should still be able to read the image data itself. I think you'd create some nasty legal issues for yourself if you were to prevent the legitimate copyright holder from accessing their images.
That said, I do agree that it'd be nice to see an open standard in use. My main question is how well a format would be able to support all current camera models (which store various amounts of information in the RAW file aside from teh sensor data) and also scale with future models? I don't know enough about the DNG format to say how it would work with this. My guess would be that it's like the TIFF format where you can freely add tags to the format. If that's the case then you could still run into the same problem with things like WB settings (or any other piece of information) even without encrypting them because different manufacturers could decide to record the information in different ways.
Thanks for the follow-up with a better link. In your original post it sounded like you were confusing encryption and proprietary formats but you've cleared that up somewhat.
I think you do need to clarify your points though since it really sounds like you're saying that all RAW files are encrypted, which isn't the case. And we're not even talking about the image information in the RAW file, we're talking about the white balance information. A lot of shooters I know don't bother with the white balance when they're shooting in RAW or they only roughly set it for if they're doing RAW+JPEG so that their low-res preview looks okay. Then they adjust the final white balance in post-processing, so it's not much of an issue.
However, there are other pieces of information that can be stored in a RAW file that can be more useful. Off the top of my head, I know some systems (such as Olympus' 4/3 system) are capable of receiving information from the lens regarding lens characteristics (barrel distortion, chromatic aberrations, etc), in theory that information could be used by the RAW processor to correct for certain lens problems. I'm not sure if it's currently being done but the potential is there. If that information were encrypted then that would only be possible in the manufacturers software (until the encryption is broken).
So I do agree that encrypting data in the RAW format is bad. I see the issue with it being proprietary as being a separate issue. Mainly, the file format is supposed to be the raw data from the sensor. This means that the format can change frequently as new models are released with different sensors. I'd certainly like it if there was a standard open format but I'm not sure how feasible it is to create a format that would continue to work with newer sensors or with camera systems that include additional information with the RAW file.
It's not a question of the data being encrypted, it's a proprietary format. Actually, it's not even a single format, it varies by manufacturer and even within models from the same manufacturer. Some manufacturers are pretty good about providing providing details about their RAW format(s) while others aren't. Your statements cast all manufacturers in the same light.
The RAW file is simply the raw data that the sensor recorded (okay, there would be some basic file header info such as EXIF headers too). Yes, some camaeras do still process the data to a certain extent (such as noise reduction) but the file is supposed to be the closest thing to an unprocessed file you can get from the camera. The intention of this is to give you the most to work with when you are editing the image later.
I would like to see cameras use a single open format. I know Adobe has been trying to push the DNG format, I'm not sure how open it is but it would at least be a single standard format.
Yeah, while I gave up on reading all the comments it sounds like the route is still being advertised but the traffic is being blocked. I don't know who RRTW uses for their connections. It doesn't really matter though if L3 is still advertising an invalid route. RRTW could be multi-homed with any number of providers but if the route to Cogent via L3 still seems the best then their routers will keep trying to send traffic that way. The only way I can think of to get around that is to set up static routes for all of Cogent's IPs to force the traffic to take other paths but that's a pain in the ass to set up.
As for why Cogent doesn't turn off the link (could just shutdown the interface vs. pulling the fibre). Well they've got a big marketing thing going now by offering free traffic for a year to anyone affected if they hook up with them. If they disable the interface then that makes the problem invisible and doesn't encourage people to switch over to them. Meanwhile it sounds like L3 may have a legitimate beef regarding how much traffic Cogent was sending them vs. how much they were sending to Cogent. If that's true than they do have the right to cancel the peering agreement and by keeping things broken they're probably hoping for that to force Cogent into paying for a connection to them.
Ah, there isn't any routing info that has to change at the modem end. The way routing works is that you don't know the whole route from end to end, it's figured out as you go.
From your computer there's only 1 path it can take, out the modem and to the ISP (unless you have multiple connections to your home network). Where to goes from there depends on the destination and how the ISP has their network setup. A pretty typical situation would be something like this:
Is the packet going to another user on the same piece of head end equipment? If so, send it back out the appropriate interface.
Is the packet going to another user on the local network but a different piece of head end equipment? If so then send it along. Depending on the network it might be direct connection or a connection to a router that figures out which piece of equipment to send it to.
Is the packet going to another user of this ISP but not on the same local network (ie, different city). Send it to the appropriate local router to send it off to the other network. Depending on the way the ISP is structured this may go through their internet connectivity or they may have their own fibre run to each city (more expensive to run but saves you in the long-run).
Is the packet going to an external IP? Send it through to your backbone routers. This may be via another city's connection (see above) depending on how the network is setup.
So, there's nothing your modem has to "know" about the routing, it just sends it to the ISP's routers for them to figure it out. If it's taken them that long to fix the problem then they probably don't have a good multi-homed setup or they have a lot of static routing that needs to be changed or Level 3 and/or Cogent are still advertising the route as valid but are blocking the traffic.
TWRR can change their routing tables all the time. They could completely change backbone providers for that matter. The only effect the end user would see could be changes in response times and transfer speeds to various sites since it's taking a new route. There can be some disruption in the initial change over until the routing change propogates but that should be fairly quick on the local network.
Well that's what I was saying. You can't just make up a new TLD and expect it to go anywhere. Likewise, you can't (easily) setup a new dns system. You could do it but it's not backward compatible. Now part of the point is that it wouldn't be backward compatible (so that we can ditch all the problems with the old system) but it's not likely to be an easy sell. Primarily, because you can't really give out your url and reliably expect people to be able to find your page. You'd have to tell people "go to foo.bar, oh wait, are you using newdns?" And while that's all good and well for the bunch of us that know how to configure BIND and run your own local DNS server (ie, local to your autonomous system, one in which you have control over) it doesn't work too well for joe-blow end-user who doesn't have a clue what DNS even is.
Now if you simply wanted to add new TLDs and get around the endless process that the current system has ("there'll be some new TLDs real soon now, no really!") then I suppose you could setup a set of root (., dot domain) servers which would direct requests to old TLDs (like.com) to the appopriate server, and new TLDs to your own servers. This gets the backward compatibility working and all that's required is to dl a new root servers config file. However, the big problem is still one of adoption. If no one goes to it then you can't really expect people to get to your url unless their DNS server is using the new root servers file.
Generally speaking, it's not a good idea to create a dueling standard, especially is the previous one is well entrenched. Also, there really needs to be a tangible improvement. Just telling people "yeah well the current system is all messed up so we decided to do our own" doesn't really work all that well. I agree that the system is messed up. It would be really nice if we could come up with a set of TLDs that are strictly comercial, and a set that are strictly non-comercial. If it were properly regulated, then a company could take the comercial ones, but they wouldn't be allowed to use the company name (or trademarks) in the non-comercial namespace. Trademark law would still apply to obvious rip-off domains (like microsoft.) but something like coke or ibm (if used as a legit abreviation or acronym) should be allowed. The real problem isn't the DNS system, it's the TLD system that the DNS system uses. Unfortunately, it seems that it's largely being driven by comercial interests. If non-comercial interests could get some decent representation then things could be improved (but it's hard to undo a lot of the damage). I mean let's look at the original system:
.com should be for comercial, period. .org should be for organizations (only non-profit even?), period. .net doesn't really make a lot of sense but should be reserved for networking type things (say ISPs, hosting services, etc), period.
Network Solutions really shouldn't be encouraging people to expand their net presence by registering all 3. What's really missing from all of this though is a TLD for personal sites. It obviously wasn't a consideration when this was all created but now it's really a large gap (I plan to register a domain I'll probably go for.com or.org simply because of recognition issues). Country codes should also have been required from the start (even for the US!). If everything was broken done by country and then by category (.com,.net, etc) then things would be a lot better off. There's also the.int domain for international stuff. Imagine how much better this would all be! If I went to www.microsoft.com.int I should be presented with choice of country (and also info about the company itself) say. Then I might want the Canadian stuff, so that would take me to www.microsoft.com.ca, it makes perfect sense, but because it wasn't thought of originally (and certainly never really adhered to), it doesn't work, and we can't really turn back time and remake it.
So you mean that the people of Nicaragua are happy that THEIR revolutionary movement (the Sandinistas) were able to overcome the US-backed Dictatorship and the US-backed counter-revolutionary (Contra, remember the Iran-Contra scandal?) forces?
You're right, they are happier, but no thanks to anything done by the US. When the Sandinistas held free elections (complete with international obeservation to make sure they were fair) the US fought against that, both by claiming that the elections were rigged when the observers said they weren't and by funding forces to attempt to overthrow the government.
You might want to check your facts before putting your foot in your mouth.
As for Vietnam, the French ran it with an iron fist and then the Vietnamese rose up against a dictatorship. The fact that it was a communist uprising is purely coincidental. The reason why the people rose up was because they wanted their own self-rule. There happened to be a group of communists that wanted to get rid of the French so they got support. The US came in to protect France's imperialistic interests but were able to cover this up under the guise of "protecting democracy".
Yeah, but don't you get into trouble because of the fact that all the TLDs are really sub-domains of the . (dot) domain? While it would be possible to create new TLDs, it wouldn't work because when you go to do a DNS lookup you start at the . and then go to.com. then foobar.com. So you'd be able to add a TLD in your Bind config but you aren't going to get anywhere from it (unless you just get everyone to use your server as their DNS server but then you're looking at lots of overhead).
While I think that this is a great idea and would really help out the problem, I don't think it's enforcable.
What would probably happen is that it would just be driven underground more. Instead of buying the domain, you give them money in order to "encourage" them to give it up (thus leaving it open for you).
Plus, you get into issues of international law. ICANN can say that domain names can't be sold, but they don't have the authority to enact a law. I suppose that where they do control who the registrars are, that they could "blacklist" non-compliant registrars. And, they do have ultimate authority in a domain dispute don't they? Would they have the authority to require that a registrar revoke the domains of a squatter? Perhaps this could work.
You're making the classic mistake of equating bandwidth and latency, they are 2 different (although often related) things. If I have to send 33Kbits of data over a 56K modem, then it will take 1sec (because I can only upload at 33.6kbps). However, on my 100mbps ethernet (preferably full-duplex and switched to the desktop ) that same ammount of data can be sent in milliseconds, so the latency (ping time) is significantly lower. The actual bandwidth usage though, is the same.
You might want to double check your own facts moron. The internet was under construction in the 70's. Appartently, you've never heard of a thing called Arpanet, which was the original internet, and was created by the gov't and universities. The mid-80's was when companies actually started to get on board.
Maybe you should check your own facts before shooting your mouth off.
A browser that reads text should just say nothing if no alt= text exists.. Bit stupid to use ALT=""...
But that would be incorrect HTML syntax. alt is a required attribute for images. It's there because of concerns about universal access. I know it sounds silly but it really makes sense. Some images can be (at least partially) described (such as using images to provide neat text, you can put the text into the alt tag). Others can't be easily described and should be indicated as such by using alt="".
Just because it works the other way doesn't mean it's correct. IE will display a table that isn't properly closed, but it cacks badly on netscape. IE's behaviour works but it's not correct. The same is true actually for many other things. A great example is &, it's always supposed to be written as & even when it appears in a url (such as a cgi url using get). Does it work the other way? Yes, but it's still not correct. Every web designer should know how to use free tools like http://validator.w3.org/. It's a great tool to find out if your page is Really compliant. Of anyone, you'd think that the W3c would know!
It's really a shame in some ways that the web has become so accesible. On the one hand it's a great medium for free speech. However, it's unfortunately become clogged with so many peopl that think that just because they can use frontpage that they're a web designer. Some better awareness of true web design would be nice.
Personally, my first experience with majour HTML design was working for a community net. Because we used lynx (no ppp conection, you connected to us with a terminal emulator), all of the pages had to be fully lynx-compliant. It really forced you though to develop tight coding practices. I really think that it's helped make me a better web author. Most of my pages now are content-based (rather than just glitz) and almost entirely text. The (graphical) formatting I use is all still intended to be lynx-friendly. Thing is, once you work with it for awhile, you can create Lynx-friendly code without really having to think about it. I still proof my pages but even I'm suprised at how well they turn out without any modification on my behalf. Nodens
I think we're getting bogged down with the idea of MS manipulating this poll. Do I think they did, hell yes. Why, because they have such a history of doing so.
However, lets face it, as far as the vast majority of people are concerned, Bill Gates invented the internet, the mouse, and the GUI. There's still a vast number of people that haven't used a computer, or who are just learning to use one. Their 1st exposure to all of the above is probably through Windows. As far as these people are concerned, MS is an innovative company that's being victimized by Government interference.
This isn't to say that these people are unintelligent, it's just that they don't know the background. They don't know that pretty much every thing that MS has ever come out with has either been stolen, copied, or bought from someone else. They are simply uninformed about MS's history. Plus, they've gotten huge ammounts of MS marketting fed to them. Remember that win95 ad that shows the pointer opening up the icon "the internet" and then it comes up with the window, "welcome to the internet"? If you didn't know any better, you would have thought that the internet came with win95.
Lets face it, the majority of people just don't know the background and can't be relied upon for a poll like this.
While piracy is an issue, I don't think this is the main thing that the RIAA is trying to fight. Personally, I pirate software because I don't have the money to buy it. If I couldn't get it for free I'd go without. Is this costing the software industry any money? No. Now with CDs it's slightly different. I can afford to buy some CDs but not a whole lot, so I'm very picky about what I buy. I use MP3s to get a chance to listen to new bands and actually form an opinion about them (instead of the song you hear once at a party and vaguely remember it being cool). In the 3 years I've been listening to MP3s I've bought around 50 CDs. Before then my CD collection consisted of 2 CDs. In Oct. I bought 10 CDs (at a cost of ~$250 cdn). 6 of those were CDs that I already had the full album in MP3. The other CDs were all from bands that I had heard MP3s of and gotten to like and wanted to support.
What most people don't realise is that the RIAA doesn't write music and it doesn't record music. All the RIAA members do is mass-produce, distribute, and market music. For this service they take a large portion of the proceeds from the album sales. In the old model this was necessary because otherwise you couldn't get your album out to many people. The bands/artists write the music and could produce the album by renting studio space (although record companies will usually cover this). The main barrier is then getting the album to the stores.
This is the reason why the RIAA fears MP3 (as a format) and is trying so hard to fight it. If the problem was just piracy then copy protection is the conventional solution for that problem, not destroying the format itself. With MP3 (and other digital music formats) the artists themselves can distribute using the internet (at very little cost) and not need the big record companies at all. That really starts to hurt the RIAA's profits. A copy-protected format would still allow this to happen, which is why the RIAA wants to crush digital formats entirely. Baring that, they want to be the ones with the control on the format. If they control the format then they can keep the costs (for others) so high that it wouldn't be feasible to independantly distribute albums.
I think that this is another reason why DAT (which has been mentioned a few times) was crushed. Not only can it copy CDs (flawlessly), but it also would allow for the cheap production of high quality masters. The RIAA put a stop on DAT in order to keep the cost of production high to keep indie bands down. This was still the early days of CD so a cassette release wasn't that bad at that time. Cheap DAT meant cheap production, and then cheap medium with tapes. By keeping the cost of DAT high it meant that bands would have to get a record contract to be able to gain access to professional recording equipment. I think now the threat is distribution.
Remember, pirating may take some money from the RIAA but indie distribution means that the RIAA doesn't get any money for those sales. The RIAA has always been about screwing the artists, they're still trying to do this but are using piracy as the excuse.
Slashdot Mirror
Yes, Nikon does add some obscurity by encrypting the WB setting. However, the original poster made it sound as if the entire file was being encrypted, which is a very different thing.
At best, encrypting the WB setting makes things mildly more annoying for if you're working with a 3rd party RAW converter. The reason I say this is because I know a number of photographers who don't use WB when shooting and instead handle that in post-production. Others may set the WB simply so that the JPEG preview looks reasonably close to what the final shot would look like.
That aside, let's look at what the WB setting actually does. It adjusts the colour based on the colour temperature of the lighting in order to remove any colour cast. A lot of people obsess over using custom WB settings but it's not really that big a deal. For outdoor shooting you probably want to maintain the character of the lighting so you don't want a custom WB. For instance, if you're shooting just around sunrise or sunset you want that nice yellow-orange light. Setting a custom WB will eliminate that look. Another example would be a message I saw on one of the photo lists I'm on, the poster was complaining that his snow shots didn't have the slight blue colour he wanted to make things look cold. The problem was he was using a custom white balance that was balancing the snow to be pure white instead of slightly blue. Simply using the "daylight" WB would have been the best option.
For indoor shooting you're most likely dealing with either tungsten or flourescent lighting. Both of these can usually be adjust for with standard WB settings.
So what's left? The main usefulness for custom WB settings is when you're working in mixed lighting. Say an indoor shot where you have tungsten and flourescent lighting as well as daylight coming in from windows. That's a tricky lighting situation. That's when you are going to want to set a custom WB. However, you can just as easily take a shot of your grey card (which you should be using to get an accurate WB setting in the first place). If you really want to get accurate colour you should shoot a frame of a colour chart (such as a Macbeth colour chart) and adjust off of that. In your RAW converter you can then use that shot to set your WB for the whole set of shots.
As for the DMCA, it sounds like it would set Nikon up to be in a position to sue if another company decrypts their WB information without permission but you should still be able to read the image data itself. I think you'd create some nasty legal issues for yourself if you were to prevent the legitimate copyright holder from accessing their images.
That said, I do agree that it'd be nice to see an open standard in use. My main question is how well a format would be able to support all current camera models (which store various amounts of information in the RAW file aside from teh sensor data) and also scale with future models? I don't know enough about the DNG format to say how it would work with this. My guess would be that it's like the TIFF format where you can freely add tags to the format. If that's the case then you could still run into the same problem with things like WB settings (or any other piece of information) even without encrypting them because different manufacturers could decide to record the information in different ways.
I think you do need to clarify your points though since it really sounds like you're saying that all RAW files are encrypted, which isn't the case. And we're not even talking about the image information in the RAW file, we're talking about the white balance information. A lot of shooters I know don't bother with the white balance when they're shooting in RAW or they only roughly set it for if they're doing RAW+JPEG so that their low-res preview looks okay. Then they adjust the final white balance in post-processing, so it's not much of an issue.
However, there are other pieces of information that can be stored in a RAW file that can be more useful. Off the top of my head, I know some systems (such as Olympus' 4/3 system) are capable of receiving information from the lens regarding lens characteristics (barrel distortion, chromatic aberrations, etc), in theory that information could be used by the RAW processor to correct for certain lens problems. I'm not sure if it's currently being done but the potential is there. If that information were encrypted then that would only be possible in the manufacturers software (until the encryption is broken).
So I do agree that encrypting data in the RAW format is bad. I see the issue with it being proprietary as being a separate issue. Mainly, the file format is supposed to be the raw data from the sensor. This means that the format can change frequently as new models are released with different sensors. I'd certainly like it if there was a standard open format but I'm not sure how feasible it is to create a format that would continue to work with newer sensors or with camera systems that include additional information with the RAW file.
It's not a question of the data being encrypted, it's a proprietary format. Actually, it's not even a single format, it varies by manufacturer and even within models from the same manufacturer. Some manufacturers are pretty good about providing providing details about their RAW format(s) while others aren't. Your statements cast all manufacturers in the same light.
The RAW file is simply the raw data that the sensor recorded (okay, there would be some basic file header info such as EXIF headers too). Yes, some camaeras do still process the data to a certain extent (such as noise reduction) but the file is supposed to be the closest thing to an unprocessed file you can get from the camera. The intention of this is to give you the most to work with when you are editing the image later.
I would like to see cameras use a single open format. I know Adobe has been trying to push the DNG format, I'm not sure how open it is but it would at least be a single standard format.
Yeah, while I gave up on reading all the comments it sounds like the route is still being advertised but the traffic is being blocked. I don't know who RRTW uses for their connections. It doesn't really matter though if L3 is still advertising an invalid route. RRTW could be multi-homed with any number of providers but if the route to Cogent via L3 still seems the best then their routers will keep trying to send traffic that way. The only way I can think of to get around that is to set up static routes for all of Cogent's IPs to force the traffic to take other paths but that's a pain in the ass to set up.
As for why Cogent doesn't turn off the link (could just shutdown the interface vs. pulling the fibre). Well they've got a big marketing thing going now by offering free traffic for a year to anyone affected if they hook up with them. If they disable the interface then that makes the problem invisible and doesn't encourage people to switch over to them. Meanwhile it sounds like L3 may have a legitimate beef regarding how much traffic Cogent was sending them vs. how much they were sending to Cogent. If that's true than they do have the right to cancel the peering agreement and by keeping things broken they're probably hoping for that to force Cogent into paying for a connection to them.
Ah, there isn't any routing info that has to change at the modem end. The way routing works is that you don't know the whole route from end to end, it's figured out as you go.
From your computer there's only 1 path it can take, out the modem and to the ISP (unless you have multiple connections to your home network). Where to goes from there depends on the destination and how the ISP has their network setup. A pretty typical situation would be something like this:
So, there's nothing your modem has to "know" about the routing, it just sends it to the ISP's routers for them to figure it out. If it's taken them that long to fix the problem then they probably don't have a good multi-homed setup or they have a lot of static routing that needs to be changed or Level 3 and/or Cogent are still advertising the route as valid but are blocking the traffic.
TWRR can change their routing tables all the time. They could completely change backbone providers for that matter. The only effect the end user would see could be changes in response times and transfer speeds to various sites since it's taking a new route. There can be some disruption in the initial change over until the routing change propogates but that should be fairly quick on the local network.Well that's what I was saying. You can't just make up a new TLD and expect it to go anywhere. Likewise, you can't (easily) setup a new dns system. You could do it but it's not backward compatible. Now part of the point is that it wouldn't be backward compatible (so that we can ditch all the problems with the old system) but it's not likely to be an easy sell. Primarily, because you can't really give out your url and reliably expect people to be able to find your page. You'd have to tell people "go to foo.bar, oh wait, are you using newdns?" And while that's all good and well for the bunch of us that know how to configure BIND and run your own local DNS server (ie, local to your autonomous system, one in which you have control over) it doesn't work too well for joe-blow end-user who doesn't have a clue what DNS even is.
.com) to the appopriate server, and new TLDs to your own servers. This gets the backward compatibility working and all that's required is to dl a new root servers config file. However, the big problem is still one of adoption. If no one goes to it then you can't really expect people to get to your url unless their DNS server is using the new root servers file.
.com or .org simply because of recognition issues). Country codes should also have been required from the start (even for the US!). If everything was broken done by country and then by category (.com, .net, etc) then things would be a lot better off. There's also the .int domain for international stuff. Imagine how much better this would all be! If I went to www.microsoft.com.int I should be presented with choice of country (and also info about the company itself) say. Then I might want the Canadian stuff, so that would take me to www.microsoft.com.ca, it makes perfect sense, but because it wasn't thought of originally (and certainly never really adhered to), it doesn't work, and we can't really turn back time and remake it.
Now if you simply wanted to add new TLDs and get around the endless process that the current system has ("there'll be some new TLDs real soon now, no really!") then I suppose you could setup a set of root (., dot domain) servers which would direct requests to old TLDs (like
Generally speaking, it's not a good idea to create a dueling standard, especially is the previous one is well entrenched. Also, there really needs to be a tangible improvement. Just telling people "yeah well the current system is all messed up so we decided to do our own" doesn't really work all that well. I agree that the system is messed up. It would be really nice if we could come up with a set of TLDs that are strictly comercial, and a set that are strictly non-comercial. If it were properly regulated, then a company could take the comercial ones, but they wouldn't be allowed to use the company name (or trademarks) in the non-comercial namespace. Trademark law would still apply to obvious rip-off domains (like microsoft.) but something like coke or ibm (if used as a legit abreviation or acronym) should be allowed. The real problem isn't the DNS system, it's the TLD system that the DNS system uses. Unfortunately, it seems that it's largely being driven by comercial interests. If non-comercial interests could get some decent representation then things could be improved (but it's hard to undo a lot of the damage). I mean let's look at the original system:
.com should be for comercial, period.
.org should be for organizations (only non-profit even?), period.
.net doesn't really make a lot of sense but should be reserved for networking type things (say ISPs, hosting services, etc), period.
Network Solutions really shouldn't be encouraging people to expand their net presence by registering all 3. What's really missing from all of this though is a TLD for personal sites. It obviously wasn't a consideration when this was all created but now it's really a large gap (I plan to register a domain I'll probably go for
Nodens
So you mean that the people of Nicaragua are happy that THEIR revolutionary movement (the Sandinistas) were able to overcome the US-backed Dictatorship and the US-backed counter-revolutionary (Contra, remember the Iran-Contra scandal?) forces?
You're right, they are happier, but no thanks to anything done by the US. When the Sandinistas held free elections (complete with international obeservation to make sure they were fair) the US fought against that, both by claiming that the elections were rigged when the observers said they weren't and by funding forces to attempt to overthrow the government.
You might want to check your facts before putting your foot in your mouth.
As for Vietnam, the French ran it with an iron fist and then the Vietnamese rose up against a dictatorship. The fact that it was a communist uprising is purely coincidental. The reason why the people rose up was because they wanted their own self-rule. There happened to be a group of communists that wanted to get rid of the French so they got support. The US came in to protect France's imperialistic interests but were able to cover this up under the guise of "protecting democracy".
Umm no. You could actually make a stronger case for the soviets winning WWWII than you can for the states.
Yeah, but don't you get into trouble because of the fact that all the TLDs are really sub-domains of the . (dot) domain? While it would be possible to create new TLDs, it wouldn't work because when you go to do a DNS lookup you start at the . and then go to .com. then foobar.com. So you'd be able to add a TLD in your Bind config but you aren't going to get anywhere from it (unless you just get everyone to use your server as their DNS server but then you're looking at lots of overhead).
Nodens
While I think that this is a great idea and would really help out the problem, I don't think it's enforcable.
What would probably happen is that it would just be driven underground more. Instead of buying the domain, you give them money in order to "encourage" them to give it up (thus leaving it open for you).
Plus, you get into issues of international law. ICANN can say that domain names can't be sold, but they don't have the authority to enact a law. I suppose that where they do control who the registrars are, that they could "blacklist" non-compliant registrars. And, they do have ultimate authority in a domain dispute don't they? Would they have the authority to require that a registrar revoke the domains of a squatter? Perhaps this could work.
Nodens
You're making the classic mistake of equating bandwidth and latency, they are 2 different (although often related) things. If I have to send 33Kbits of data over a 56K modem, then it will take 1sec (because I can only upload at 33.6kbps). However, on my 100mbps ethernet (preferably full-duplex and switched to the desktop ) that same ammount of data can be sent in milliseconds, so the latency (ping time) is significantly lower. The actual bandwidth usage though, is the same.
You might want to double check your own facts moron. The internet was under construction in the 70's. Appartently, you've never heard of a thing called Arpanet, which was the original internet, and was created by the gov't and universities. The mid-80's was when companies actually started to get on board.
Maybe you should check your own facts before shooting your mouth off.
But that would be incorrect HTML syntax. alt is a required attribute for images. It's there because of concerns about universal access. I know it sounds silly but it really makes sense. Some images can be (at least partially) described (such as using images to provide neat text, you can put the text into the alt tag). Others can't be easily described and should be indicated as such by using alt="".
Just because it works the other way doesn't mean it's correct. IE will display a table that isn't properly closed, but it cacks badly on netscape. IE's behaviour works but it's not correct. The same is true actually for many other things. A great example is &, it's always supposed to be written as & even when it appears in a url (such as a cgi url using get). Does it work the other way? Yes, but it's still not correct. Every web designer should know how to use free tools like http://validator.w3.org/. It's a great tool to find out if your page is Really compliant. Of anyone, you'd think that the W3c would know!
It's really a shame in some ways that the web has become so accesible. On the one hand it's a great medium for free speech. However, it's unfortunately become clogged with so many peopl that think that just because they can use frontpage that they're a web designer. Some better awareness of true web design would be nice.
Personally, my first experience with majour HTML design was working for a community net. Because we used lynx (no ppp conection, you connected to us with a terminal emulator), all of the pages had to be fully lynx-compliant. It really forced you though to develop tight coding practices. I really think that it's helped make me a better web author. Most of my pages now are content-based (rather than just glitz) and almost entirely text. The (graphical) formatting I use is all still intended to be lynx-friendly. Thing is, once you work with it for awhile, you can create Lynx-friendly code without really having to think about it. I still proof my pages but even I'm suprised at how well they turn out without any modification on my behalf. Nodens
I think we're getting bogged down with the idea of MS manipulating this poll. Do I think they did, hell yes. Why, because they have such a history of doing so.
However, lets face it, as far as the vast majority of people are concerned, Bill Gates invented the internet, the mouse, and the GUI. There's still a vast number of people that haven't used a computer, or who are just learning to use one. Their 1st exposure to all of the above is probably through Windows. As far as these people are concerned, MS is an innovative company that's being victimized by Government interference.
This isn't to say that these people are unintelligent, it's just that they don't know the background. They don't know that pretty much every thing that MS has ever come out with has either been stolen, copied, or bought from someone else. They are simply uninformed about MS's history. Plus, they've gotten huge ammounts of MS marketting fed to them. Remember that win95 ad that shows the pointer opening up the icon "the internet" and then it comes up with the window, "welcome to the internet"? If you didn't know any better, you would have thought that the internet came with win95.
Lets face it, the majority of people just don't know the background and can't be relied upon for a poll like this.
While piracy is an issue, I don't think this is the main thing that the RIAA is trying to fight. Personally, I pirate software because I don't have the money to buy it. If I couldn't get it for free I'd go without. Is this costing the software industry any money? No. Now with CDs it's slightly different. I can afford to buy some CDs but not a whole lot, so I'm very picky about what I buy. I use MP3s to get a chance to listen to new bands and actually form an opinion about them (instead of the song you hear once at a party and vaguely remember it being cool). In the 3 years I've been listening to MP3s I've bought around 50 CDs. Before then my CD collection consisted of 2 CDs. In Oct. I bought 10 CDs (at a cost of ~$250 cdn). 6 of those were CDs that I already had the full album in MP3. The other CDs were all from bands that I had heard MP3s of and gotten to like and wanted to support.
What most people don't realise is that the RIAA doesn't write music and it doesn't record music. All the RIAA members do is mass-produce, distribute, and market music. For this service they take a large portion of the proceeds from the album sales. In the old model this was necessary because otherwise you couldn't get your album out to many people. The bands/artists write the music and could produce the album by renting studio space (although record companies will usually cover this). The main barrier is then getting the album to the stores.
This is the reason why the RIAA fears MP3 (as a format) and is trying so hard to fight it. If the problem was just piracy then copy protection is the conventional solution for that problem, not destroying the format itself. With MP3 (and other digital music formats) the artists themselves can distribute using the internet (at very little cost) and not need the big record companies at all. That really starts to hurt the RIAA's profits. A copy-protected format would still allow this to happen, which is why the RIAA wants to crush digital formats entirely. Baring that, they want to be the ones with the control on the format. If they control the format then they can keep the costs (for others) so high that it wouldn't be feasible to independantly distribute albums.
I think that this is another reason why DAT (which has been mentioned a few times) was crushed. Not only can it copy CDs (flawlessly), but it also would allow for the cheap production of high quality masters. The RIAA put a stop on DAT in order to keep the cost of production high to keep indie bands down. This was still the early days of CD so a cassette release wasn't that bad at that time. Cheap DAT meant cheap production, and then cheap medium with tapes. By keeping the cost of DAT high it meant that bands would have to get a record contract to be able to gain access to professional recording equipment. I think now the threat is distribution.
Remember, pirating may take some money from the RIAA but indie distribution means that the RIAA doesn't get any money for those sales. The RIAA has always been about screwing the artists, they're still trying to do this but are using piracy as the excuse.