Right, this is what I meant by having Google Voice call one of your numbers to patch you through. To the recipient, it appears to come from your Google Voice phone number.
There is also an iPhone app to automate this, but I don't think it works with Google Voice (just Grand Central). Hopefully Google will come out with something official or the app gets updated.
I don't use it that often, because I typically don't care if someone gets my real number. However, when calling car dealers it's invaluable.
It is nice that you can turn off Call Presentation now. I wish, as I did with Grand Central, that the level of configurability would get way higher. Things like having certain people's calls go through without the Call Presentation thing.
It would also be nice if the system was complex enough to understand voice commands in addition to the numbers. The biggest pain I have is answering a call on my iPhone requires changing over to keypad mode every time to hit '1'. However, it pays for itself when I manage to avoid a call that I really didn't want to take.
Another sweet feature: the contacts are now held in your Google Contacts stuff (shared with Gmail, which caused me a few initial problems), so that you can sync that up with the iPhone (and I presume other phones somehow) as well. It's really becoming a Googlefied world.
On the plus side, they did add some settings if you're concerned about this. Under Advanced Settings for each phone, you can now control whether or not it requires a PIN to access voicemail.
With Grand Central, devices listed as 'mobile' just got special treatment, but now it's a little finer grained.
I'm not really sure how else they could handle this, besides just eliminating the PIN-less voicemail and account control features entirely or having the default as off with big warnings about the boogeymen who will get you.
It's the same service as Grand Central, which I've been using for 2-3 years now.
The basic idea is that you can hide all of your various phone numbers behind your Google Voice number. People call it and all of your phones (or the ones you have configured for that caller or at that time of day) will ring. Whichever one you pick up gets the call, and you will be told the person's name and given the choice to actually answer or bounce them to voicemail.
On the other side, you can use the web interface to have Google Voice call one of your phones and connect you with any phone number you give it. This is free, except for international calls. I don't use this too often, but it helps when you don't want people to find out one of your 'real' phone numbers.
The best part is that you can control incoming calls essentially with a spam filter. When people call you they have to state their name (the first time), which plays when you answer their calls. You can decide to bounce certain numbers straight to voicemail every time or give them a 'this number is not in service' message.
Google Voice added the following features that I like:
- Voicemails are transcribed, not very well but you can usually get the jist quickly without listening - SMS is now forwarded as well, which was pretty much the major short-coming of Grand Central.
Overall, I really like it, and the service quality has been quite good. The main thing is that it is not a phone service in itself, but something you use with other phone services.
Since they're using one of Sun's modular datacenters that is actually on the Sun campus, I would imagine that they got some financial incentives / support from Sun for all of this.
The X4500 is EOL as you mention, although it was still sold a few months back. It lives on as the X4540, which really isn't that different; the main thing is it's moved to a newer Opteron processor type and is a fair bit cheaper. So they didn't really miss out on anything.
It's kind of interesting to me that they went this route, as opposed to a bunch of servers talking to a bunch of storage separately. This seems to be an exact use case for the X4500-type system, which as far as I'm aware is pretty unique.
The only advertising that I've noticed is the little banner that appears in the upper right corner, which in my experience is always for the same product shown during the in-video ads.
I guess Hulu's recommended video listings could be considered ads as well, since they're intended to drive you to other Hulu video offerings rather than just watching whatever afterwards, as you would be more likely to do with Boxee.
Yes, all patents expire. Because of the date on which this patent (#5579517) was filed, it will expire 17 years after the issue date (Nov 26, 1996). So approximately November 2013.
However, it seems clear that Microsoft, as with most companies in this position, will continue to develop new derivative work that can be patented in turn. This will allow them to continue to constraint the filesystem under patent as it will be implemented in 2013, which may or may not matter depending on how the world works in the distant future.
Why should Amazon increase the fee when the publisher and author are not adding any value? Rather Amazon has added the value here at their own expense.
By having TTS capabilities, they have not eliminated the audiobook market but have combined the regular market with the audiobook market into one cohesive (e-book) market. This may reduce redundant sales (when someone buys multiple formats of the same basic product), but may also prevent market inefficiencies such as a person desiring one version of the product but not being able to find it due to stocking issues.
As it stands with regular books, a given person may have to buy the same basic product twice in order to have it in both a written and audio form. This may be worth it because the audiobook form has a special value above and beyond the book version, but fundamentally they are paying twice for the basic content. By enhancing the book, the publisher gets to charge more and this is the financial incentive to undertake the work. The only reason why they are able to charge more is because of the scarcity of the ability to create this product; most people do not have recording studios and access to persons with good speaking voices.
The Kindle (and similar technology) has removed that scarcity, and so the need to produce audiobooks will decline. However, it seems unlikely that it will entirely disappear, as there will be a difference between artificial and natural readers for some time to come.
Just because you were able to successfully exploit market demands for a while doesn't mean that you should be able to do forever regardless of technological progress. Should automobiles have a built-in horse whip tax in order to keep that industry afloat?
I think the major point is that the same rules should not apply to Microsoft because they have a monopoly position in the OS market (or at least a lot of people see them that way).
Microsoft has typically been felt to utilize a monopoly in one area to establish a monopoly in another area, and I think one instance of this is the rapid way that Internet Explorer took over the web browser market. This in turn can be utilized to establish de facto control over web standards, which in turn can be utilized to influence web development, and so on.
With a company with offerings in as many categories as Microsoft has, it's very easy for them to tie strong, popular products to weak or new ones. To an extent this is their prerogative, as any manufacturer can make their products work best in their own environment. The issue is that Microsoft has historically gone beyond 'works best' to 'only works with' and that they have such a size advantage that it has become anti-competitive.
I agree that it would be absurd to have an OS without a web browser at this point, but calling for equal treatment of different OS players would require that the playing field was level, which is not the case.
I don't know what you think universities cost today, but $40,000 per year is not uncommon. Whether or not it is worth it is another matter, but you can find somewhere willing to charge in that range easily.
The university that I went to is now up to $46,000 per year, although only $34,000 of that is for tuition and another $1,000 in required fees.
Nosferatu isn't a very good example, because it actually was ordered to be destroyed as a result of a suit from Bram Stoker's estate. Luckily not all of the copies were destroyed.
I agree, 'enterprise' solutions are often more trouble than they're worth.
There are a lot of solutions that look good on paper and then turn out to be serious pains in the ass in practice, or have a failure mode that is actually worse than the common method (but perhaps less likely).
One of my biggest gripes about systems administration is that there are all of these solutions to make life easier, but a lot of them are basically traps if you are running a large scale operation. Take centralized authentication for instance. Makes life a lot easier, reduces the possibility of mistakes due to manual intervention. Yet if that service is unavailable, it could bring the entire infrastructure down and require a non-trivial solution.
So then you fall back on the old method of manually updating/etc/passwd, which is also a super pain. You make something to automate the process, push out updates to systems as necessary (or use something like puppet or cfengine).
I guess my point is, this stuff has been thought through often, but the technology isn't ubiquitous to handle the problems. Companies have their own internal glue to deal with these type of problems, and it keeps getting re-invented over and over again.
I agree to an extent, but I also think that not all of these sites will survive their re-implementation periods in the face of better-designed competitors. Flickr, for instance, is internally a mess. I presume part of this is due to poor initial implementation, but its further compounded by a need to Yahooize it at every level.
I presume Twitter will encounter a mass exodus at some point, as its users are likely to be very keen to move on to the next big (and possibly more reliable) thing.
Every time a site is down, you run the risk of irretrievably losing a portion of your users. Once you get enough bad will going, you don't even have to have failures; just having a reputation as not being reliable can be enough.
OK, lets explore this. If I was to log to syslog, only the ErrorLog supports it. In order to do this with an AccessLog, I would have to use the piped log output feature to route to a script that I write which in turn writes to syslog for Apache.
This is exactly the sort of bespoke stuff I'm referring to. Why should this need to be implemented 1,000 times at company after company to accomplish the exact same thing?
I agree that there is a trade-off here. If you spend too much, take too long and aren't 'agile' enough then your site will be old news by the time you get it out of the gate. No one cares, and its all pointless.
On the other hand, if you don't spend any time worrying about the future you will be totally unprepared if you reach your goal of user interest. Then the site doesn't work sufficiently well to retain users, no one cares and its all pointless.
I think part of the overall issue is that while there are numerous frameworks and reusable components to ease development, most of them don't really add anything when it comes to future maintainability. In my experience, they often detract.
So much of what it takes to run a site properly is bespoke and closely guarded, unlike the vast development resources out there for the taking. Its good in a way (it keeps me employed), but its also a waste of time. Ideally we should be solving new problems, not wasting time having Apache rotate logs without restarting for the 1,000,000th time.
Twitter's infrastructure is notoriously poorly thought out, and I sort of doubt they employed any systems administrators (or service engineers, or operations engineers, or whatever) up until recently.
I think the barrier to entry from an engineering standpoint has been lowered such that you can more easily make a site that appears to be pretty decent and attracts an audience. What is often missing is the behind-the-scenes work which ensures that the service is:
- Deployed properly, with testing and staging environments that actually mirror production. - Fault-tolerant at every practical level. This gets expensive, so you see datacenter failures take down large swaths of sites who don't have multiple locations. - Constantly monitored, including performance metrics, to find issues quickly or ever before they happen.
This is the kind of work that always seems to take a back seat to development due to resource constraints, but it really needs to occur in tandem with the development process.
If you don't design a site from the ground up to be redundant and highly performing, its pretty difficult to flip a switch and make it that way later. Which is basically what Twitter has found out. Whether or not this mentality is taking over the Interworld is another story though.
Metropolis was in the public domain, but became copyrighted again in 1998.
However, keep in mind that even if the original film itself were in the public domain today, acquiring a copy of this would be quite difficult. The restored versions are covered under separate copyright, as is the music. Since this particular film has had a lot of (positive) restoration work performed, the value of having its original pre-restoration version in the public domain is pretty minor.
The additional footage hasn't been seen by anyone in my lifetime, which is why this discovery is such a big deal.
Its possible that you saw one of the longer versions of the film, as it has had several major restorations over the years after new stock is found. Some of these attempt to pad out the missing scenes with still production photos.
Then there is the question of which intertitles were used. If they were in English, they aren't the original ones, and thus they might be on screen for a different amount of time (or there could be more or less of them than in German).
There is also the problem of the correct speed to play this film back at. While modern films are standardized at 24fps, films of this era were generally not intended to be played at that speed. Although the 'standard' silent speed is 16fps, this could actually vary between films. I've even seen some talk that Metropolis was designed to take advantage of hand cranking and thus was intended to have a variable frame rate at different parts.
Depending on what speed your projectionist used, the movie's runtime could vary wildly.
Kino has already stated that they intend to include this footage on their new DVD/BluRay release of the movie in 2009. However, its not totally clear yet if it will be incorporated into the film or as a supplement.
For one thing, OSX has much more frequent releases (including the occasional 10.x.x release that actually changes things around) which spread the risk of large changes out over a period of time.
Whereas Microsoft seems to want to make each version of Windows a radical re-invention that takes 8 years to brew (and then mostly excludes much talked about features anyway).
Consider that unlocking a phone is just the first step. Frequencies differ between carriers and countries. For instance, a 3G ATT phone uses the 1900MHz band while a 3G T-Mobile phone uses 1700MHz.
Then theres CDMA carriers like Verizon and Sprint, which are totally incompatible with the GSM carriers.
In a nutshell cell phones suck and there are a lot of reasons why phones wind up in landfills.
There was even quite a bit of press at the time about how Microsoft had decided to dump NVidia in favor of ATI after being so buddy-buddy with them on the original Xbox.
Who knows. Yahoo certainly doesn't use Yahoo Mail internally, they use Exchange.
And while I use GMail, it certainly does have some failings that I might not want to bet the entire company on. Its IMAP support for one is a
A company isn't guaranteed to use its own products, especially if their products aren't targeted towards large companies. I did loathe Exchange though.
Slashdot Mirror
Right, this is what I meant by having Google Voice call one of your numbers to patch you through. To the recipient, it appears to come from your Google Voice phone number.
There is also an iPhone app to automate this, but I don't think it works with Google Voice (just Grand Central). Hopefully Google will come out with something official or the app gets updated.
I don't use it that often, because I typically don't care if someone gets my real number. However, when calling car dealers it's invaluable.
It is nice that you can turn off Call Presentation now. I wish, as I did with Grand Central, that the level of configurability would get way higher. Things like having certain people's calls go through without the Call Presentation thing.
It would also be nice if the system was complex enough to understand voice commands in addition to the numbers. The biggest pain I have is answering a call on my iPhone requires changing over to keypad mode every time to hit '1'. However, it pays for itself when I manage to avoid a call that I really didn't want to take.
Another sweet feature: the contacts are now held in your Google Contacts stuff (shared with Gmail, which caused me a few initial problems), so that you can sync that up with the iPhone (and I presume other phones somehow) as well. It's really becoming a Googlefied world.
On the plus side, they did add some settings if you're concerned about this. Under Advanced Settings for each phone, you can now control whether or not it requires a PIN to access voicemail.
With Grand Central, devices listed as 'mobile' just got special treatment, but now it's a little finer grained.
I'm not really sure how else they could handle this, besides just eliminating the PIN-less voicemail and account control features entirely or having the default as off with big warnings about the boogeymen who will get you.
It's the same service as Grand Central, which I've been using for 2-3 years now.
The basic idea is that you can hide all of your various phone numbers behind your Google Voice number. People call it and all of your phones (or the ones you have configured for that caller or at that time of day) will ring. Whichever one you pick up gets the call, and you will be told the person's name and given the choice to actually answer or bounce them to voicemail.
On the other side, you can use the web interface to have Google Voice call one of your phones and connect you with any phone number you give it. This is free, except for international calls. I don't use this too often, but it helps when you don't want people to find out one of your 'real' phone numbers.
The best part is that you can control incoming calls essentially with a spam filter. When people call you they have to state their name (the first time), which plays when you answer their calls. You can decide to bounce certain numbers straight to voicemail every time or give them a 'this number is not in service' message.
Google Voice added the following features that I like:
- Voicemails are transcribed, not very well but you can usually get the jist quickly without listening
- SMS is now forwarded as well, which was pretty much the major short-coming of Grand Central.
Overall, I really like it, and the service quality has been quite good. The main thing is that it is not a phone service in itself, but something you use with other phone services.
Since they're using one of Sun's modular datacenters that is actually on the Sun campus, I would imagine that they got some financial incentives / support from Sun for all of this.
The X4500 is EOL as you mention, although it was still sold a few months back. It lives on as the X4540, which really isn't that different; the main thing is it's moved to a newer Opteron processor type and is a fair bit cheaper. So they didn't really miss out on anything.
It's kind of interesting to me that they went this route, as opposed to a bunch of servers talking to a bunch of storage separately. This seems to be an exact use case for the X4500-type system, which as far as I'm aware is pretty unique.
The only advertising that I've noticed is the little banner that appears in the upper right corner, which in my experience is always for the same product shown during the in-video ads.
I guess Hulu's recommended video listings could be considered ads as well, since they're intended to drive you to other Hulu video offerings rather than just watching whatever afterwards, as you would be more likely to do with Boxee.
Yes, all patents expire. Because of the date on which this patent (#5579517) was filed, it will expire 17 years after the issue date (Nov 26, 1996). So approximately November 2013.
However, it seems clear that Microsoft, as with most companies in this position, will continue to develop new derivative work that can be patented in turn. This will allow them to continue to constraint the filesystem under patent as it will be implemented in 2013, which may or may not matter depending on how the world works in the distant future.
Why should Amazon increase the fee when the publisher and author are not adding any value? Rather Amazon has added the value here at their own expense.
By having TTS capabilities, they have not eliminated the audiobook market but have combined the regular market with the audiobook market into one cohesive (e-book) market. This may reduce redundant sales (when someone buys multiple formats of the same basic product), but may also prevent market inefficiencies such as a person desiring one version of the product but not being able to find it due to stocking issues.
As it stands with regular books, a given person may have to buy the same basic product twice in order to have it in both a written and audio form. This may be worth it because the audiobook form has a special value above and beyond the book version, but fundamentally they are paying twice for the basic content. By enhancing the book, the publisher gets to charge more and this is the financial incentive to undertake the work. The only reason why they are able to charge more is because of the scarcity of the ability to create this product; most people do not have recording studios and access to persons with good speaking voices.
The Kindle (and similar technology) has removed that scarcity, and so the need to produce audiobooks will decline. However, it seems unlikely that it will entirely disappear, as there will be a difference between artificial and natural readers for some time to come.
Just because you were able to successfully exploit market demands for a while doesn't mean that you should be able to do forever regardless of technological progress. Should automobiles have a built-in horse whip tax in order to keep that industry afloat?
I think the major point is that the same rules should not apply to Microsoft because they have a monopoly position in the OS market (or at least a lot of people see them that way).
Microsoft has typically been felt to utilize a monopoly in one area to establish a monopoly in another area, and I think one instance of this is the rapid way that Internet Explorer took over the web browser market. This in turn can be utilized to establish de facto control over web standards, which in turn can be utilized to influence web development, and so on.
With a company with offerings in as many categories as Microsoft has, it's very easy for them to tie strong, popular products to weak or new ones. To an extent this is their prerogative, as any manufacturer can make their products work best in their own environment. The issue is that Microsoft has historically gone beyond 'works best' to 'only works with' and that they have such a size advantage that it has become anti-competitive.
I agree that it would be absurd to have an OS without a web browser at this point, but calling for equal treatment of different OS players would require that the playing field was level, which is not the case.
I don't know what you think universities cost today, but $40,000 per year is not uncommon. Whether or not it is worth it is another matter, but you can find somewhere willing to charge in that range easily.
The university that I went to is now up to $46,000 per year, although only $34,000 of that is for tuition and another $1,000 in required fees.
Nosferatu isn't a very good example, because it actually was ordered to be destroyed as a result of a suit from Bram Stoker's estate. Luckily not all of the copies were destroyed.
I agree, 'enterprise' solutions are often more trouble than they're worth.
There are a lot of solutions that look good on paper and then turn out to be serious pains in the ass in practice, or have a failure mode that is actually worse than the common method (but perhaps less likely).
One of my biggest gripes about systems administration is that there are all of these solutions to make life easier, but a lot of them are basically traps if you are running a large scale operation. Take centralized authentication for instance. Makes life a lot easier, reduces the possibility of mistakes due to manual intervention. Yet if that service is unavailable, it could bring the entire infrastructure down and require a non-trivial solution.
So then you fall back on the old method of manually updating /etc/passwd, which is also a super pain. You make something to automate the process, push out updates to systems as necessary (or use something like puppet or cfengine).
I guess my point is, this stuff has been thought through often, but the technology isn't ubiquitous to handle the problems. Companies have their own internal glue to deal with these type of problems, and it keeps getting re-invented over and over again.
I agree to an extent, but I also think that not all of these sites will survive their re-implementation periods in the face of better-designed competitors. Flickr, for instance, is internally a mess. I presume part of this is due to poor initial implementation, but its further compounded by a need to Yahooize it at every level.
I presume Twitter will encounter a mass exodus at some point, as its users are likely to be very keen to move on to the next big (and possibly more reliable) thing.
Every time a site is down, you run the risk of irretrievably losing a portion of your users. Once you get enough bad will going, you don't even have to have failures; just having a reputation as not being reliable can be enough.
OK, lets explore this. If I was to log to syslog, only the ErrorLog supports it. In order to do this with an AccessLog, I would have to use the piped log output feature to route to a script that I write which in turn writes to syslog for Apache.
This is exactly the sort of bespoke stuff I'm referring to. Why should this need to be implemented 1,000 times at company after company to accomplish the exact same thing?
I agree that there is a trade-off here. If you spend too much, take too long and aren't 'agile' enough then your site will be old news by the time you get it out of the gate. No one cares, and its all pointless.
On the other hand, if you don't spend any time worrying about the future you will be totally unprepared if you reach your goal of user interest. Then the site doesn't work sufficiently well to retain users, no one cares and its all pointless.
I think part of the overall issue is that while there are numerous frameworks and reusable components to ease development, most of them don't really add anything when it comes to future maintainability. In my experience, they often detract.
So much of what it takes to run a site properly is bespoke and closely guarded, unlike the vast development resources out there for the taking. Its good in a way (it keeps me employed), but its also a waste of time. Ideally we should be solving new problems, not wasting time having Apache rotate logs without restarting for the 1,000,000th time.
Twitter's infrastructure is notoriously poorly thought out, and I sort of doubt they employed any systems administrators (or service engineers, or operations engineers, or whatever) up until recently.
I think the barrier to entry from an engineering standpoint has been lowered such that you can more easily make a site that appears to be pretty decent and attracts an audience. What is often missing is the behind-the-scenes work which ensures that the service is:
- Deployed properly, with testing and staging environments that actually mirror production.
- Fault-tolerant at every practical level. This gets expensive, so you see datacenter failures take down large swaths of sites who don't have multiple locations.
- Constantly monitored, including performance metrics, to find issues quickly or ever before they happen.
This is the kind of work that always seems to take a back seat to development due to resource constraints, but it really needs to occur in tandem with the development process.
If you don't design a site from the ground up to be redundant and highly performing, its pretty difficult to flip a switch and make it that way later. Which is basically what Twitter has found out. Whether or not this mentality is taking over the Interworld is another story though.
Metropolis was in the public domain, but became copyrighted again in 1998.
However, keep in mind that even if the original film itself were in the public domain today, acquiring a copy of this would be quite difficult. The restored versions are covered under separate copyright, as is the music. Since this particular film has had a lot of (positive) restoration work performed, the value of having its original pre-restoration version in the public domain is pretty minor.
The additional footage hasn't been seen by anyone in my lifetime, which is why this discovery is such a big deal.
Its possible that you saw one of the longer versions of the film, as it has had several major restorations over the years after new stock is found. Some of these attempt to pad out the missing scenes with still production photos.
Then there is the question of which intertitles were used. If they were in English, they aren't the original ones, and thus they might be on screen for a different amount of time (or there could be more or less of them than in German).
There is also the problem of the correct speed to play this film back at. While modern films are standardized at 24fps, films of this era were generally not intended to be played at that speed. Although the 'standard' silent speed is 16fps, this could actually vary between films. I've even seen some talk that Metropolis was designed to take advantage of hand cranking and thus was intended to have a variable frame rate at different parts.
Depending on what speed your projectionist used, the movie's runtime could vary wildly.
Kino has already stated that they intend to include this footage on their new DVD/BluRay release of the movie in 2009. However, its not totally clear yet if it will be incorporated into the film or as a supplement.
For one thing, OSX has much more frequent releases (including the occasional 10.x.x release that actually changes things around) which spread the risk of large changes out over a period of time.
Whereas Microsoft seems to want to make each version of Windows a radical re-invention that takes 8 years to brew (and then mostly excludes much talked about features anyway).
These sites probably did base their reviews on incomplete code, and its probably the same thing being shipped to stores on CDs right now.
Consider that unlocking a phone is just the first step. Frequencies differ between carriers and countries. For instance, a 3G ATT phone uses the 1900MHz band while a 3G T-Mobile phone uses 1700MHz.
Then theres CDMA carriers like Verizon and Sprint, which are totally incompatible with the GSM carriers.
In a nutshell cell phones suck and there are a lot of reasons why phones wind up in landfills.
Back in my day we hadn't even discovered fire.
There was even quite a bit of press at the time about how Microsoft had decided to dump NVidia in favor of ATI after being so buddy-buddy with them on the original Xbox.
Who knows. Yahoo certainly doesn't use Yahoo Mail internally, they use Exchange.
And while I use GMail, it certainly does have some failings that I might not want to bet the entire company on. Its IMAP support for one is a
A company isn't guaranteed to use its own products, especially if their products aren't targeted towards large companies. I did loathe Exchange though.