One of my cats is 18 and (not to be too biased) one of the smartest cat's I've seen out of those I've known. I know he can easily recognize himself in a mirror and that he understands the concept of reflection well. He will often watch me from around the corner in a large hall mirror I have outside my office. If I call him from the office he will look at me in the mirror to see that I am motioning to him, then walk to where I am, not towards the reflection. He will also often clean in front of the mirror taking pauses to look at himself before moving on to another area. When done he'll leave for a more comfortable place to sleep.
My other cat ignored the mirror completely however he also never flipped out at his own, or his partner's reflections like he would when an unknown cat came into view. I can only infer that at some basic level he understood what a reflection was too.
What surprises me more is the limited list they assigned to mammals they thought were capable of this.
I tend to believe that modern communication and gaming communities killed the puzzle style game. Few people are 'nobel' enough to stick it out in a puzzle game without input from the outside. However these days when you run into a room in a game and you can't find the key, or open the lock you can just pop online and search the local forum for the answer. Presto... and you're through. Since the puzzles were supposed to be the challenge in the game being able to just get the answers readily from a community kills much of the challenge since those people get bored with that genre ("oh that game? yeah... I solved it in 5 hours... I'm not buying the next one because it was too easy"). The few who do quietly work their way through the entire game challenge by challenge are not enough of a user base to make the market viable for game makers.
I remember the first Alone in the Dark.. I was addicted to that game and there really wasn't much of a community to walk you through it. You, and maybe the 3-4 other people you knew personally who owned it had to work through it and you didn't want to share too much because there is competition inside your social circle. When you can post an answer on a forum for strangers... and in turn anonymously get an answer from strangers on a forum... a large percentage of people will cave and look it up.
In a short and highly unscientific test I searched for 4 sites I am involved which, all of which rank quite well on all the other search engines, especially when you use certain terms (like business names)
Of the 4 sites... 3 appear to not be indexed at all and the 4th came up with the wrong image beside it on page 6.
Methinks it was a little too early to launch and get this much press.
I do a lot of work for trade show events and you should see what a big-city arena-type venue charges to drop an internet connection to a booth on a show floor. $1000 for a 3-5 day even is not at all uncommon. (granted typically at higher speeds than mentioned in this article and wired access... but for 5 days)
When you have a captive client base you can get away with amazing things.
My mother is a typical late 60's web user... she has a handful of site she likes to visit and not much more. She has memorized the basic functions I taught her years ago and she's happy with that.
Recently I upgraded her FF2 to FF3 and taught her how to use the new address bar and bookmarking / search functionality. She nailed it in 2-3 minutes and was looking up sites in her history with ease. I was back there a couple days ago and sure enough she has already bookmarked a dozen new sites and raves about how much easier she finds the internet now. (you'd think they had redesigned the entire internet... which in essence is what a browser upgrade can do for you)
To me that right there outlines one of the reasons FF3 is going to produce another large spike in new users. Get what you want easily and with less hassle.
Not going to happen anywhere near that quickly. The act of lifting your arm up to touch a screen vs. moving to a mouse will lead to a whole new world of strain injuries not to mention that it's just not efficient.
Tablet style devices will become much more common and of course won't use a mouse but for the standard keyboard based system (which the article says isn't going anywhere) the ergonomics of a KB/Mouse combination just isn't going to be replaced any time soon by a touch screen. (augmented, yes... replaced... no.)
Scrolling up and down unnecessary pages of code attempting to decipher what is going on is even more expensive. Clean and formatted code is obviously is highly important but there is also a value in code density. Striking a balance is important and to me having 20 lines of lone { and }'s on the screen swings that balance too far into the 'useless' category. I don't compact code to any extreme but I do try to avoid generally "blank" one character lines inside connected constructs.
"} else {" is not unreasonable to me at all. It keeps the construct together and if your eye has to scan to see something 2 characters away then you're sitting way too damn close to the screen:)
"The researchers then asked whether the structural connections of the brain in fact shape its dynamic activity, Sporns said. The study examined the brains of five human participants who were imaged using both fMRI and DSI techniques to compare how closely the brain activity observed in the fMRI mapped to the underlying fiber networks.
"It turns out they're quite closely related," Sporns said. "We can measure a significant correlation between brain anatomy and brain dynamics. This means that if we know how the brain is connected we can predict what the brain will do."
Phrenology might be on the come back! Quick, get me my skull maps!
If an aftermarket upgrade part on your car causes the engine to blow up do you blame the original manufacturer of the car?
It's not feasible to expect a framework to catch every possible problem that could arise from a poor implementation of that framework. You'd like to catch most of them but there will always be someone or something that can fubar it in ways you never imagined.
Reading from the memory cache is faster, yes. But in surfing dozens of pages an hour, not everything stays in memory. And swap-file is not RAM. And even the RAM cache is usually of teh html, not the internally parsed version.
That's a skewed argument as outside of the initial page load a typical user would be repeatedly accessing your site during their period of interest and not be heading elsewhere mid session. The files in question would be recently cached with a high potential of being still in memory. Even if it had degraded to the disk cache it's still local and much quicker to read.
Disk-based cached file is faster than downloading a 10K file, but not faster than downloading an additional 10K mid-stream. With no connection over-head, downloading 10K at full speed on a broadband connection is 1/100 of a second. The file-system isn't always that fast to locate a file. Standard game, try to copy one 500Mb file, and see that it takes a minute. Try to copy 500 1Mb files, and watch it take ten minutes. Finding and opening files can be really slow. And the problem gets worse when this becomes a principle and you've got three external files.
The users you need to optimize for are specifically the ones on slower connections, not faster. 10kb of data in 1/100th of a second is a sustained rate of ~1 Mb/second, or about optimum at the highest level of service my Cable ISP offers. That's a fairly optimistic expectation of your average client's connection and of your server to be able to push out that level of traffic to multiple users at the same time. For a user on a typical 768kbit/s DSL connection that's more like 0.1s for that 10k of JS. Or 0.5s for 50k of JS. A web browser can find and read a file out of its cache in much less than 0.1s. And if the file was accessed only 30 seconds ago on a previous page load odds are it's sitting in RAM and not on disk. If the browser is set to check a file on every load then you have the network latency of the call to check the file however that is still well under 0.1s and especially the 0.5s for the theoretically larger file.
For personal interest I just did up a small app that reads into memory 1000 random files out of my Firefox cache directory (a set of ~2500 files) The average time to locate and read a file into memory averages out to be about 0.012s per. I'm sure that in comparison that is quite slow to an actual browser considering this was a quickly kludged and non-optimized experiment.
As for code maintenance, you can have it separated for development, but you shouldn't be forcing the user to suffer because you want to do less work. Package it up again so that the separation doesn't make it all the way to the end user. What if that were an actual technical requirement of a weird project? Like for a legitimate reason? Would you not be able to do it? Would it be more work? It's not for me, I've got enough platforms and techniques to get around this issue without exposing the end user.
It's as simple as a server side include, in fact most of my pages are compentized and only assembled on request. The point of moving things like common JS and CSS externally is to allow the browser's cache to hold some of the data to save the client the overhead of retrieving it again across different page loads and some server bandwidth which will impact all clients visiting your site.
You mentioned in another post about inlining image data. You can do that with data URLS on supported browsers (read: without work-arounds, anything that's not 5 >= IE <= 7). Since IE8 is finally re-adding native support for inline images it may well become more common in the future. So where is the tipping point? Would you inline 30k of images on every page load as well? 100k? 500k? I suspect the crux of the issue is in the size of the content involved. There comes a point when the additional drag on the client and server exceeds the benefits, especially on clients using slower connectio
While I don't agree with the position in the initial post I will clarify it for you. What he's saying is it's faster to d/load the extra 10k embedded inside the HTML file (which you're going to have to open/read anyhow) than retrieving it as a separate file via the local cache and then opening that.
Something that may be affecting the differing results you two are seeing is that the call to check if a file has been modified is browser and user settings dependent. This is most clearly demonstrated in IE's Temp Files settings where you can choose how frequently it checks for updates to cached files.
I don't really get how this breaches confidentiality agreements. The most Google would see is a call from the client for the JS file in question. That's less information than if you are using Google Ads or even driving a call to Google Maps out of your site since I don't believe it contains referrer information as well. They could tell who is loading that file when but since no actual client data is being transferred there isn't much at risk.
The external file requires yet another hit to the server, and everything involved therein. It almost never makes any sense. In a nutshell it reduces overall load on your server, it saves your clients on slower connections from re-downloading the same content over and over, and it allows you to do centralized code updates without having to remember every file that uses that routine. Myself I tend to use a combination of in-line and centralized code. Anything that gets used more than once goes in the common file and anything that is specific to one page goes inline on that page.
Even as a locally cached file, on a broadband connection, downloading the extra 10K is typically faster than opening and reading the locally cached file! This part I really don't understand. When it is a disk based call it is faster than downloading however the difference becomes even greater once you consider that most of the time (outside initial load) the client is already going to be holding the page in its memory cache as well. Reading existing data from RAM is exponentially faster than pushing it through the network and re-parsing it each time.
That's what I thought of immediately. Get OpenOffice, make a spreadsheet, throw on some column filters and you have the best quicky database you can find with practically zero investment. Everything else is overkill.
I still have my 2 Commodore PET's and associated cassette and disk drives along with all my code from back then. Every couple years I fire one up and load a program or two for fun. Last time was about a year ago and all worked like a charm.
IMO watch any 80's series of that genre and you'll find they were generally crap (with a couple notable exceptions) The actors have matured, the concept will mature, but most importantly the expectations of the audience have changed dramatically. Today TV shows have the type of plot and character depth you didn't even get in movies in the 80's. I think any new movie would (I'd hope) reflect the standards of today's audience.
RDA to play MacGyver. If Harrison Ford can reprise Indy, RDA can easily do a movie or two. I *personally* don't think the franchise concept has enough gas in it to spawn a huge following and a long series of movies, etc. with a new face in the role. Most of the big fans really just want to re-live the late 80's... see MacGyver stop global annihilation with chewing gum and a paperclip a few more times... and put a nice wrapper on the whole series. No better way to do that than with RDA.
Duchovny and Anderson in a new X-Files movie.
Ford reviving Indy
Willis reviving Die Hard
Stalone in new Rambo and Rocky movies
It's a no brainer.. all making good money. RDA has to do MacGyver and Selleck has to do Magnum.
Simple reason anyone buys anything. The quality was there and the price was right.
There is also the aspect of co-operation between nations in space travel to offset the load on any one organization/country. These days we are used to that concept with the ISS, however it existed to a lesser degree even back in the early shuttle days.
Primarily though I'd guess that NASA threw out a spec and proposal request and MDA said "Arms? We can build arms. We've been reaching for beer for years and sometimes them suckers are a long way away". Once the shuttle arm worked well, MDA was in NASA's good books.
"I've had hundreds more problems with FF rendering than IE rendering. That's because I start with IE, and build FF second. It's been that way for developers since the dawn of time. The primary browser is easy, the secondaries are not. You get to pick the one that you want primary."
And I am the reverse... I develop in FF first and make sure it renders there fine, then I also find that it works well in Opera and Safari 99% of the time right off the cuff... then I go back and fix poor lagging IE by working around it's rendering issues. Heck in IE7 MS ponied up and re-worked many of their known issues admitting they had been lagging behind the standards on their own blogs.
"But I'll say again what I've been saying for well over a decade. The day the W3C makes a browser that correctly meets each and every one of their standards, then I'll point to IE and others and say that they suck. Until then, it's the W3C that sucks. I too can invent standards that are difficult to follow."
So because the FAA doesn't build airplanes they shouldn't set guidelines for manufacturers? And because the FCC doesn't produce any TV programs they shouldn't define guidelines what can be broadcast? How about because GM is the largest auto manufacturer we let them decide how cars should be built and what guidelines they should adhere to? Then all of the other manufacturers could just follow along.
Can you imagine what the Internet would be like if we didn't have organized bodies attempting to set a standard everyone can agreed on? Can we say "free for all"? Can you imagine a world without standards? No broadcast TV signal standards, no adherence to a wireless spectrum rules, heck my telephone probably wouldn't even work outside my local calling area. The W3C is attempting to bring the same form of organization to the web and set one common set of rules that make things inter-operate at an expected level. That is in no way a bad goal or a fruitless effort. Nor should it be ignored.
Slashdot Mirror
... or the sun-spots.
One of my cats is 18 and (not to be too biased) one of the smartest cat's I've seen out of those I've known. I know he can easily recognize himself in a mirror and that he understands the concept of reflection well. He will often watch me from around the corner in a large hall mirror I have outside my office. If I call him from the office he will look at me in the mirror to see that I am motioning to him, then walk to where I am, not towards the reflection. He will also often clean in front of the mirror taking pauses to look at himself before moving on to another area. When done he'll leave for a more comfortable place to sleep.
My other cat ignored the mirror completely however he also never flipped out at his own, or his partner's reflections like he would when an unknown cat came into view. I can only infer that at some basic level he understood what a reflection was too.
What surprises me more is the limited list they assigned to mammals they thought were capable of this.
I tend to believe that modern communication and gaming communities killed the puzzle style game. Few people are 'nobel' enough to stick it out in a puzzle game without input from the outside. However these days when you run into a room in a game and you can't find the key, or open the lock you can just pop online and search the local forum for the answer. Presto... and you're through. Since the puzzles were supposed to be the challenge in the game being able to just get the answers readily from a community kills much of the challenge since those people get bored with that genre ("oh that game? yeah... I solved it in 5 hours... I'm not buying the next one because it was too easy"). The few who do quietly work their way through the entire game challenge by challenge are not enough of a user base to make the market viable for game makers.
I remember the first Alone in the Dark.. I was addicted to that game and there really wasn't much of a community to walk you through it. You, and maybe the 3-4 other people you knew personally who owned it had to work through it and you didn't want to share too much because there is competition inside your social circle. When you can post an answer on a forum for strangers... and in turn anonymously get an answer from strangers on a forum... a large percentage of people will cave and look it up.
In a short and highly unscientific test I searched for 4 sites I am involved which, all of which rank quite well on all the other search engines, especially when you use certain terms (like business names)
Of the 4 sites... 3 appear to not be indexed at all and the 4th came up with the wrong image beside it on page 6.
Methinks it was a little too early to launch and get this much press.
I do a lot of work for trade show events and you should see what a big-city arena-type venue charges to drop an internet connection to a booth on a show floor. $1000 for a 3-5 day even is not at all uncommon. (granted typically at higher speeds than mentioned in this article and wired access... but for 5 days)
When you have a captive client base you can get away with amazing things.
We could call it the "Ba-dum-bum!" mod level :) (I have yet to figure out how to type a cymbal crash)
Surprising to me... I have it on 4 systems here and it hasn't hiccuped once yet under heavy use.
I can sum much of this up with one example.
My mother is a typical late 60's web user... she has a handful of site she likes to visit and not much more. She has memorized the basic functions I taught her years ago and she's happy with that.
Recently I upgraded her FF2 to FF3 and taught her how to use the new address bar and bookmarking / search functionality. She nailed it in 2-3 minutes and was looking up sites in her history with ease. I was back there a couple days ago and sure enough she has already bookmarked a dozen new sites and raves about how much easier she finds the internet now. (you'd think they had redesigned the entire internet... which in essence is what a browser upgrade can do for you)
To me that right there outlines one of the reasons FF3 is going to produce another large spike in new users. Get what you want easily and with less hassle.
Not going to happen anywhere near that quickly. The act of lifting your arm up to touch a screen vs. moving to a mouse will lead to a whole new world of strain injuries not to mention that it's just not efficient.
Tablet style devices will become much more common and of course won't use a mouse but for the standard keyboard based system (which the article says isn't going anywhere) the ergonomics of a KB/Mouse combination just isn't going to be replaced any time soon by a touch screen. (augmented, yes... replaced... no.)
2 Commodore PET's in the closet and not going anywhere :) Both still work beautifully.
Scrolling up and down unnecessary pages of code attempting to decipher what is going on is even more expensive. Clean and formatted code is obviously is highly important but there is also a value in code density. Striking a balance is important and to me having 20 lines of lone { and }'s on the screen swings that balance too far into the 'useless' category. I don't compact code to any extreme but I do try to avoid generally "blank" one character lines inside connected constructs.
:)
"} else {" is not unreasonable to me at all. It keeps the construct together and if your eye has to scan to see something 2 characters away then you're sitting way too damn close to the screen
Looks like someone goofed...
http://www.heliumraven.com/ninareisercase
RIP Nina.
"The researchers then asked whether the structural connections of the brain in fact shape its dynamic activity, Sporns said. The study examined the brains of five human participants who were imaged using both fMRI and DSI techniques to compare how closely the brain activity observed in the fMRI mapped to the underlying fiber networks.
"It turns out they're quite closely related," Sporns said. "We can measure a significant correlation between brain anatomy and brain dynamics. This means that if we know how the brain is connected we can predict what the brain will do."
Phrenology might be on the come back! Quick, get me my skull maps!
If an aftermarket upgrade part on your car causes the engine to blow up do you blame the original manufacturer of the car? It's not feasible to expect a framework to catch every possible problem that could arise from a poor implementation of that framework. You'd like to catch most of them but there will always be someone or something that can fubar it in ways you never imagined.
Yay! I got one! A FF3 download complete.
I feel like one of those parents that beats the crap out of other parents on Christmas Eve for the last Furby in Walmart.
Reading from the memory cache is faster, yes. But in surfing dozens of pages an hour, not everything stays in memory. And swap-file is not RAM. And even the RAM cache is usually of teh html, not the internally parsed version.
That's a skewed argument as outside of the initial page load a typical user would be repeatedly accessing your site during their period of interest and not be heading elsewhere mid session. The files in question would be recently cached with a high potential of being still in memory. Even if it had degraded to the disk cache it's still local and much quicker to read.
Disk-based cached file is faster than downloading a 10K file, but not faster than downloading an additional 10K mid-stream. With no connection over-head, downloading 10K at full speed on a broadband connection is 1/100 of a second. The file-system isn't always that fast to locate a file. Standard game, try to copy one 500Mb file, and see that it takes a minute. Try to copy 500 1Mb files, and watch it take ten minutes. Finding and opening files can be really slow. And the problem gets worse when this becomes a principle and you've got three external files.
The users you need to optimize for are specifically the ones on slower connections, not faster. 10kb of data in 1/100th of a second is a sustained rate of ~1 Mb/second, or about optimum at the highest level of service my Cable ISP offers. That's a fairly optimistic expectation of your average client's connection and of your server to be able to push out that level of traffic to multiple users at the same time. For a user on a typical 768kbit/s DSL connection that's more like 0.1s for that 10k of JS. Or 0.5s for 50k of JS. A web browser can find and read a file out of its cache in much less than 0.1s. And if the file was accessed only 30 seconds ago on a previous page load odds are it's sitting in RAM and not on disk. If the browser is set to check a file on every load then you have the network latency of the call to check the file however that is still well under 0.1s and especially the 0.5s for the theoretically larger file.
For personal interest I just did up a small app that reads into memory 1000 random files out of my Firefox cache directory (a set of ~2500 files) The average time to locate and read a file into memory averages out to be about 0.012s per. I'm sure that in comparison that is quite slow to an actual browser considering this was a quickly kludged and non-optimized experiment.
As for code maintenance, you can have it separated for development, but you shouldn't be forcing the user to suffer because you want to do less work. Package it up again so that the separation doesn't make it all the way to the end user. What if that were an actual technical requirement of a weird project? Like for a legitimate reason? Would you not be able to do it? Would it be more work? It's not for me, I've got enough platforms and techniques to get around this issue without exposing the end user.
It's as simple as a server side include, in fact most of my pages are compentized and only assembled on request. The point of moving things like common JS and CSS externally is to allow the browser's cache to hold some of the data to save the client the overhead of retrieving it again across different page loads and some server bandwidth which will impact all clients visiting your site.
You mentioned in another post about inlining image data. You can do that with data URLS on supported browsers (read: without work-arounds, anything that's not 5 >= IE <= 7). Since IE8 is finally re-adding native support for inline images it may well become more common in the future. So where is the tipping point? Would you inline 30k of images on every page load as well? 100k? 500k? I suspect the crux of the issue is in the size of the content involved. There comes a point when the additional drag on the client and server exceeds the benefits, especially on clients using slower connectio
While I don't agree with the position in the initial post I will clarify it for you. What he's saying is it's faster to d/load the extra 10k embedded inside the HTML file (which you're going to have to open/read anyhow) than retrieving it as a separate file via the local cache and then opening that.
Something that may be affecting the differing results you two are seeing is that the call to check if a file has been modified is browser and user settings dependent. This is most clearly demonstrated in IE's Temp Files settings where you can choose how frequently it checks for updates to cached files.
That's what I thought of immediately. Get OpenOffice, make a spreadsheet, throw on some column filters and you have the best quicky database you can find with practically zero investment. Everything else is overkill.
I still have my 2 Commodore PET's and associated cassette and disk drives along with all my code from back then. Every couple years I fire one up and load a program or two for fun. Last time was about a year ago and all worked like a charm.
IMO watch any 80's series of that genre and you'll find they were generally crap (with a couple notable exceptions) The actors have matured, the concept will mature, but most importantly the expectations of the audience have changed dramatically. Today TV shows have the type of plot and character depth you didn't even get in movies in the 80's. I think any new movie would (I'd hope) reflect the standards of today's audience.
RDA to play MacGyver. If Harrison Ford can reprise Indy, RDA can easily do a movie or two. I *personally* don't think the franchise concept has enough gas in it to spawn a huge following and a long series of movies, etc. with a new face in the role. Most of the big fans really just want to re-live the late 80's... see MacGyver stop global annihilation with chewing gum and a paperclip a few more times... and put a nice wrapper on the whole series. No better way to do that than with RDA.
Duchovny and Anderson in a new X-Files movie.
Ford reviving Indy
Willis reviving Die Hard
Stalone in new Rambo and Rocky movies
It's a no brainer.. all making good money. RDA has to do MacGyver and Selleck has to do Magnum.
Simple reason anyone buys anything. The quality was there and the price was right.
There is also the aspect of co-operation between nations in space travel to offset the load on any one organization/country. These days we are used to that concept with the ISS, however it existed to a lesser degree even back in the early shuttle days.
Primarily though I'd guess that NASA threw out a spec and proposal request and MDA said "Arms? We can build arms. We've been reaching for beer for years and sometimes them suckers are a long way away". Once the shuttle arm worked well, MDA was in NASA's good books.
- a pleased Canuck
"I've had hundreds more problems with FF rendering than IE rendering. That's because I start with IE, and build FF second. It's been that way for developers since the dawn of time. The primary browser is easy, the secondaries are not. You get to pick the one that you want primary."
And I am the reverse... I develop in FF first and make sure it renders there fine, then I also find that it works well in Opera and Safari 99% of the time right off the cuff... then I go back and fix poor lagging IE by working around it's rendering issues. Heck in IE7 MS ponied up and re-worked many of their known issues admitting they had been lagging behind the standards on their own blogs.
"But I'll say again what I've been saying for well over a decade. The day the W3C makes a browser that correctly meets each and every one of their standards, then I'll point to IE and others and say that they suck. Until then, it's the W3C that sucks. I too can invent standards that are difficult to follow."
So because the FAA doesn't build airplanes they shouldn't set guidelines for manufacturers? And because the FCC doesn't produce any TV programs they shouldn't define guidelines what can be broadcast? How about because GM is the largest auto manufacturer we let them decide how cars should be built and what guidelines they should adhere to? Then all of the other manufacturers could just follow along.
Can you imagine what the Internet would be like if we didn't have organized bodies attempting to set a standard everyone can agreed on? Can we say "free for all"? Can you imagine a world without standards? No broadcast TV signal standards, no adherence to a wireless spectrum rules, heck my telephone probably wouldn't even work outside my local calling area. The W3C is attempting to bring the same form of organization to the web and set one common set of rules that make things inter-operate at an expected level. That is in no way a bad goal or a fruitless effort. Nor should it be ignored.