AI doth never prosper, what's the reason? If it prosper, none dare call it AI.
We've been seeing this since Turing and the Turing test. If it is an unsolved, apparently open ended problem, it must be an AI application. As soon as someone figures out how to do it, it wasn't really AI after all, just an obscure, but somehow also mainstream, technique to be worked out. They move the goal posts and say AI is now over there in that other unsolved problem.
You guys are talking past each other. But in essence that is how MySQL is set up. They will license you a proprietary source copy, or you can use the Open Source one under a differing set of terms. Of course the packages themselves are somewhat different too.
If you really believe all of those tools will come up with the same answer, you are the fool. They are incredibly useful. But it still takes expert interpretation to turn them into useful information.
Try this on for size: I want a program that says "Hello World!". That is the entire requirement. Give me a time and cost estimate to get it done.
Hint: how many languages and operating environments are there? Don't forget output to video display, to printers, to plotters, or to audio output.
Even a really easy thing can be complicated. There is no single correct answer. It is a range, and you pick the one, in your opinion, that seems most probable. The tools lead you there, but you decide. And then you manage to it, and live or die by your decision.
Plans can change. Just don't change it without telling anyone. Telling them BEFORE the change, that is. If you do it after they'll figure it out on their own. During the investigation / witch hunt.
Agile is great for quickly verifying your project is moving in the right direction. I'm just wary of it being use as a substitute for a robust system design. Users might be happy with the basic functionality they see, but they may not be in a good position to see the big picture of long term maintenance or system architecture. If adding a new function turns into a massive hack to get it to work, you aren't doing them a favor by being highly responsive up front.
But by no means are the two mutually exclusive. They can easily coexist and be better for it. I would never expect people to know everything up front. I just want them to know something, and to understand pulling rabbits out of the hat isn't free.
It isn't about waterfall methodology vs anything else at all. It is more about Hitting the target, and trying to keep the target from moving more than absolutely necessary.
I would personally prefer that all requirements be firm and well known in advance. Then have a firm spec developed, then have a fixed and approved design, etc. It probably is not going to be possible or practical to do business that way, but it would be nice.
But what IS possible is to avoid having a new and unvetted requirement suddenly land in the middle of the development cycle and have the developers start reacting to it. They run off and build some really cool widget that the customer really likes, but no one knows exactly how it fits into the scheme of everything else that has been tested to spec. So now you have code that kinda does something we think is useful but we aren't sure how or where to put it. Which strongly implies you need to go back into system design and maybe back to the architecture and "fix it" so the widget can stay. Then comes the recoding of already accomplished work. All of which costs time and money. BAD BAD BAD practice. It is an invitation to chaos. You never know where you are going or when you are done. But the costs keep coming and time rolls on.
As I once said to a team, "Lets say you are flying out to Hawaii from California. Half way, you decide you'd rather go to Tahiti. A while later you change course for Fiji. What are the chances you'll end up at any of those places, vs just swimming in mid-Pacific?"
If the customer has something to add, let them say it. Write it down, look it over. Figure out how it might change the existing design. GET A COST FOR IT. Then, find a good time in the development cycle to integrate a controlled change to a NEW BASELINE that includes the new widget. The design works, it is testable, maintainable, the users are happy, and the cost is known and understood before you commit to it.
And I'd be wary about calling something a "release" when you know it doesn't meet the requirements. "Prototype", "release candidate", "testing demo" maybe, but not Release. Never give the impression that it is something the users should be using when you know they should not. But that is a semantics matter for another time.
In addition, be be careful with your requirements, specifications, and testing.
Your users and customers (two related but often slightly different groups) are supposed to come up with the requirements, but often they are clueless on what they need. So you will often need to help them with suggested feasible solutions. However, the ultimate decision on what is REQUIRED is theirs. Just be sure to help them with the difference between required vs nice to have vs "you have got to be dreaming". The budget and time estimate is based on the requirement.
ONCE THE REQUIREMENTS ARE LOCKED DOWN you do not accept changes to them. Any changes go into a NEW requirement that will be harmonized with the old one at a later date. Think of it like a train leaving the station. No new passengers get on, none of the old ones jump off, except under controlled conditions. If the users want to change the requirement, tell them to get on the next train. As the PM, you decide when the new stuff can be included into the old AND HOW MUCH IT WILL COST TO DO IT. Never let them think it will be "free".
Getting a good estimate from the written requirement is tough. Trying to determine Function Points and lines of code and complexity and speed of development is a serious art form. Get good people and go over it a lot, from different angles. If you are lucky, this project is similar enough to past projects that you won't plant the seeds of destruction at this stage. You need to be sure you can really live with the cost and time estimate you give them. DO NOT ASSUME BEST CASE just because it look "easy". Too many people do. DON'T JUST DOUBLE EVERYTHING unknown. that is just wasteful. If you have serious unknowns, do some risk reduction explorations to be sure you do know what you are talking about (or at least plan to do them so you will know when the time comes).
The best specifications are testable. And the Tests should be written at about the time the specs are written. A Requirement might say "full color display". A Specification might say, "display in at least six colors, including white, black, red, green, blue, cyan, magenta, and yellow". Guess what the acceptance test is going to look for? It should be as Unambiguous as possible. This is where team work is good. Don't let the designer write the specs and the tests. Too much chance for hidden assumptions to creep in.
Which reminds me, be sure to explicitly lay out the overall software design, all the modules, all the interfaces, and subject them all to thorough rigorous Reviews. Too many otherwise good projects die from unstated assumptions that lurk under the surface. The coders are so anxious to get started they forget to examine where they are and where they are going vs the tools and skills available. They never see the iceberg until too late.
Please do your best not to become another "out of control software project".
So what is or who are the United States? Its' citizens or the government? or both or something else?
In the context of the Constitution, "the United States" are or is the federal government. It seems to be plural or singular in different spots, but the term always includes the federal government. The term might also be referring to the individual states as a collection as well (cf. "the several states", "each of them", etc.).
What is war?
Not directly defined in the Constitution, but implied to be some kind of martial action intended to overthrow or displace the constitutional federal or state government.
Who are United States' enemies?
A generally political question. In context, it would be someone making war on the United States, or perhaps someone about to, but since that is a difficult thing to judge (allies come and go), it isn't in the definition.
That was a one time aberration and should not reflect on his character as a whole. The money was in the freezer only because the cupboard was already stuffed full, and there was no time to get to the storage container or make a flight to Grand Cayman.
...by interfering in an investigation, which is, of course, illegal.
Did she interfere? How?
Did she destroy or alter or manufacture evidence, withhold or alter testimony, order investigators to drop leads or to slow their work or even provide false leads? No. She asked the prosecutor's boss if he could reduce some charges. Oh wait... she promised to ask the prosecutor's boss to reduce (not withdraw) some charges.
So where did the interference come from, and where did it go?
"aid and comfort" is not just making a phone call and saying nice things about someone.
There is a long tradition of the definition, and it means something a lot closer to handing over weapons, transporting troops, feeding and clothing them, or anything of that nature. You know, actual work toward making real war. Not just having political sentiments that make some radio-blatherhead get an upset tum-tum.
The only one I can think of off the top of my head is "Cold Cash" Jefferson from New Orleans.
I object to this scurrilous and unfounded attack on the good name of a fine representative from a great city and a great state. They called him William "Dollar Bill" Jefferson.
Turns out there was a good reason Harmon needed the help. She and Pelosi had some bad blood going back a few years. When Pelosi got the Speakership, Harmon did not get the committee chair, even though she was the senior member. Traditionally she would have, but the Speaker can override tradition. Harmon must have known it was probable, and was looking for an outside push in her favor.
Her trading a favor for a favor isn't treason, and I doubt it is even a real crime of any kind. She never offered to make a vote or even introduce legislation or hire anyone. She just said she'd do her job to help a particular person, in exchange for someone helping her. That is known as politics. Unfortunately, she was doing politics with an unsavory group in an unsavory way. And it was made public. Very Bad Luck.
No, we want to help these guys get the fine paid off. You do realize that not paying for "damages" is terms for being put back into prison, right? The property they "stole" was imaginary, but the money they have to pay has to be real.
The money in an electronic funds transfer is no more real than the "imaginary" property you refer to. So why is one any harder to deal with than the other? Its all just bits and bytes. Ignoring one should be of no more consequence that ignoring the other, right?
These guys are heroes. They're putting their asses on the line for our right to copy - may it be equal to everyone else's - end copyright now.
Yes, part of this is about your right to copy. Personal copies, and fair use a part of this debate. But don't kid yourself. The serious issue here is not the right to copy. It is the right to copy and then distribute copies in a way that infringes on or negates the right of the creator to distribute or sell the product. In effect it is about the right to steal.
If you were distributing bread out of the back of a bakery without the baker's permission, on the theory that everyone has a right to bread, you might have plenty of takers. But you would have not a moral leg to stand on. It would not matter that you have a cool technological way to take the bread off the shelves, nor that the baker charges more that you think the bread should cost, nor that the bread would have gone uneaten had you not taken it and given it away. Nor would any other argument that gives you the power to distribute the bread in any way or for any reason that is against the wishes of its creator.
People would simply recognize it for the theft that it is. And that would be true, any way you slice it.
This entire story stinks of a distinct lack of personal responsiblity. As far as analogies go, think of it as someone who abandons a property for months on end, allowing the grass to grow high, paint to begin peeling off the siding, and animals to take up residence in the living room. The owner returns to said derelict property and is shocked to find a family of raccoons nesting in his lounger.
You insensitve clod! My home only looks that way to match the neighbors!
You think it is a coincidence your coworkers usually work in dank and dimly lit cubicles with no human contact? That they startle whenever anyone approaches? That they always bitch about the poor conditions at the rare and always uncomfortably awkward staff meetings, but no one ever actually attempts to change it? You're seeing it *EVERY DAY*
The whole point of a good snare is to put it in place without being noticed or felt. I'd like to think Time Warner will be very reasonable and rational about placing the limits. Put them where only 1, or 2, or 10% of the customers ever exceed them. 3+ standard deviations above the mean, etc. Something easily rationalized to a technically minded consumer.
But then the bean counters and MBAs will roll in. No matter where the limits are placed, they will try to find a way to squeeze out more profit and provide less service. Profit optimization. More and more people will find they have somehow tripped one of the new magic "penalty" clauses that will pop up every month. As more and more video and entertainment apps come along, its guaranteed that downloads and download volume will increase over time. Suddenly, you are an "excessive" user, and the jacked up rates apply.
The cost to (to the company) do so will decrease, but the price to the consumer will be driven by corporate desire for profit. And seeing as cable company internet service is nearly a monopoly, what are the chances they will drop prices as their own costs drop?
CADIE AI was base ball mad. Had the fever and had it bad; Just to root for the home town crew, Ev'ry sou CADIE blew. On a Saturday, her young beau Called to (see (if (she'd like to go))), To (see (a show)) but Miss CADIE said, "No, I'll tell you what you can do."...
Nahh. NPR is biased liberal to a large degree, but mostly in the features coverage. The news is pretty much straight off the wires field reporting.
If you want REAL left wing wild eyed stuff, listen to Pacifica radio news. They have core stations in New York, Washington, LA, and Berkeley. Just like Fox, but hitting you from the "left progressive" stance.
I can't watch Fox at home because I'd end up breaking my new plasma screen. But I can listen to Pacifica in the car because I can't really hit the radio buried in the dash.
Reporter bias, editor bias, assignment bias, publisher bias. Take your choice.
The only people not biased are people that don't have any background in the subject. And they are worthless too, since they don't understand what they are seeing and hearing.
Which leads us to how Entertainment Tonight covers political campaigns. Or maybe thats ABCNNBCBS Faux sensati-celebri-news. Hard to tell anymore.
We use Outlook/Exchange. It had a message recall button, but the function wasn't enabled. Which meant about once a week you could see a message with a followup THAT ASKED THE READER if the previous message could be recalled. Even if you said yes (after reading the mistaken message of course) the bad message did not disappear.
This worked great a flag for screwed up mail to be read first. Thanks Microsoft!
How likely is it that if a programmer shows a user some code, and the feedback is the code is too slow, that the user will be satisfied with a 2:1 or a 4:1 speedup?
2:1 is probably only just noticeable, assuming it isn't an actual timed test. Anything that highly depends on user responsiveness (i.e. gaming and simulations) needs pretty dramatic pickups before the user will categorically agree it is better.
Even the more reasonable ones will want a "meaningful" increase. So the time saved has to be enough that they could do something useful with it. i.e. shoot another bad guy, beat the market to a good deal, go out for a smoke break, get home an hour earlier, etc.
Slashdot Mirror
AI doth never prosper, what's the reason? If it prosper, none dare call it AI.
We've been seeing this since Turing and the Turing test. If it is an unsolved, apparently open ended problem, it must be an AI application. As soon as someone figures out how to do it, it wasn't really AI after all, just an obscure, but somehow also mainstream, technique to be worked out. They move the goal posts and say AI is now over there in that other unsolved problem.
You guys are talking past each other. But in essence that is how MySQL is set up. They will license you a proprietary source copy, or you can use the Open Source one under a differing set of terms. Of course the packages themselves are somewhat different too.
If you really believe all of those tools will come up with the same answer, you are the fool. They are incredibly useful. But it still takes expert interpretation to turn them into useful information.
Try this on for size: I want a program that says "Hello World!". That is the entire requirement. Give me a time and cost estimate to get it done.
Hint: how many languages and operating environments are there? Don't forget output to video display, to printers, to plotters, or to audio output.
Even a really easy thing can be complicated. There is no single correct answer. It is a range, and you pick the one, in your opinion, that seems most probable. The tools lead you there, but you decide. And then you manage to it, and live or die by your decision.
It looks like Guardian has finally been uncovered. Everybody act really friendly, no fast moves to the on/off switch.
Plans can change. Just don't change it without telling anyone. Telling them BEFORE the change, that is. If you do it after they'll figure it out on their own. During the investigation / witch hunt.
Agile is great for quickly verifying your project is moving in the right direction. I'm just wary of it being use as a substitute for a robust system design. Users might be happy with the basic functionality they see, but they may not be in a good position to see the big picture of long term maintenance or system architecture. If adding a new function turns into a massive hack to get it to work, you aren't doing them a favor by being highly responsive up front.
But by no means are the two mutually exclusive. They can easily coexist and be better for it. I would never expect people to know everything up front. I just want them to know something, and to understand pulling rabbits out of the hat isn't free.
It isn't about waterfall methodology vs anything else at all. It is more about Hitting the target, and trying to keep the target from moving more than absolutely necessary.
I would personally prefer that all requirements be firm and well known in advance. Then have a firm spec developed, then have a fixed and approved design, etc. It probably is not going to be possible or practical to do business that way, but it would be nice.
But what IS possible is to avoid having a new and unvetted requirement suddenly land in the middle of the development cycle and have the developers start reacting to it. They run off and build some really cool widget that the customer really likes, but no one knows exactly how it fits into the scheme of everything else that has been tested to spec. So now you have code that kinda does something we think is useful but we aren't sure how or where to put it. Which strongly implies you need to go back into system design and maybe back to the architecture and "fix it" so the widget can stay. Then comes the recoding of already accomplished work. All of which costs time and money. BAD BAD BAD practice. It is an invitation to chaos. You never know where you are going or when you are done. But the costs keep coming and time rolls on.
As I once said to a team, "Lets say you are flying out to Hawaii from California. Half way, you decide you'd rather go to Tahiti. A while later you change course for Fiji. What are the chances you'll end up at any of those places, vs just swimming in mid-Pacific?"
If the customer has something to add, let them say it. Write it down, look it over. Figure out how it might change the existing design. GET A COST FOR IT. Then, find a good time in the development cycle to integrate a controlled change to a NEW BASELINE that includes the new widget. The design works, it is testable, maintainable, the users are happy, and the cost is known and understood before you commit to it.
And I'd be wary about calling something a "release" when you know it doesn't meet the requirements. "Prototype", "release candidate", "testing demo" maybe, but not Release. Never give the impression that it is something the users should be using when you know they should not. But that is a semantics matter for another time.
In addition, be be careful with your requirements, specifications, and testing.
Your users and customers (two related but often slightly different groups) are supposed to come up with the requirements, but often they are clueless on what they need. So you will often need to help them with suggested feasible solutions. However, the ultimate decision on what is REQUIRED is theirs. Just be sure to help them with the difference between required vs nice to have vs "you have got to be dreaming". The budget and time estimate is based on the requirement.
ONCE THE REQUIREMENTS ARE LOCKED DOWN you do not accept changes to them. Any changes go into a NEW requirement that will be harmonized with the old one at a later date. Think of it like a train leaving the station. No new passengers get on, none of the old ones jump off, except under controlled conditions. If the users want to change the requirement, tell them to get on the next train. As the PM, you decide when the new stuff can be included into the old AND HOW MUCH IT WILL COST TO DO IT. Never let them think it will be "free".
Getting a good estimate from the written requirement is tough. Trying to determine Function Points and lines of code and complexity and speed of development is a serious art form. Get good people and go over it a lot, from different angles. If you are lucky, this project is similar enough to past projects that you won't plant the seeds of destruction at this stage. You need to be sure you can really live with the cost and time estimate you give them. DO NOT ASSUME BEST CASE just because it look "easy". Too many people do. DON'T JUST DOUBLE EVERYTHING unknown. that is just wasteful. If you have serious unknowns, do some risk reduction explorations to be sure you do know what you are talking about (or at least plan to do them so you will know when the time comes).
The best specifications are testable. And the Tests should be written at about the time the specs are written. A Requirement might say "full color display". A Specification might say, "display in at least six colors, including white, black, red, green, blue, cyan, magenta, and yellow". Guess what the acceptance test is going to look for? It should be as Unambiguous as possible. This is where team work is good. Don't let the designer write the specs and the tests. Too much chance for hidden assumptions to creep in.
Which reminds me, be sure to explicitly lay out the overall software design, all the modules, all the interfaces, and subject them all to thorough rigorous Reviews. Too many otherwise good projects die from unstated assumptions that lurk under the surface. The coders are so anxious to get started they forget to examine where they are and where they are going vs the tools and skills available. They never see the iceberg until too late.
Please do your best not to become another "out of control software project".
Because Folk went off to school and got all educated. When it came back, it never did fit in with the other kids again.
But Bluegrass will hang out, just for old time's sake.
So what is or who are the United States? Its' citizens or the government? or both or something else?
In the context of the Constitution, "the United States" are or is the federal government. It seems to be plural or singular in different spots, but the term always includes the federal government. The term might also be referring to the individual states as a collection as well (cf. "the several states", "each of them", etc.).
What is war?
Not directly defined in the Constitution, but implied to be some kind of martial action intended to overthrow or displace the constitutional federal or state government.
Who are United States' enemies?
A generally political question. In context, it would be someone making war on the United States, or perhaps someone about to, but since that is a difficult thing to judge (allies come and go), it isn't in the definition.
That was a one time aberration and should not reflect on his character as a whole. The money was in the freezer only because the cupboard was already stuffed full, and there was no time to get to the storage container or make a flight to Grand Cayman.
Did she interfere? How?
Did she destroy or alter or manufacture evidence, withhold or alter testimony, order investigators to drop leads or to slow their work or even provide false leads? No. She asked the prosecutor's boss if he could reduce some charges. Oh wait... she promised to ask the prosecutor's boss to reduce (not withdraw) some charges.
So where did the interference come from, and where did it go?
"aid and comfort" is not just making a phone call and saying nice things about someone.
There is a long tradition of the definition, and it means something a lot closer to handing over weapons, transporting troops, feeding and clothing them, or anything of that nature. You know, actual work toward making real war. Not just having political sentiments that make some radio-blatherhead get an upset tum-tum.
The only one I can think of off the top of my head is "Cold Cash" Jefferson from New Orleans.
I object to this scurrilous and unfounded attack on the good name of a fine representative from a great city and a great state. They called him William "Dollar Bill" Jefferson.
Turns out there was a good reason Harmon needed the help. She and Pelosi had some bad blood going back a few years. When Pelosi got the Speakership, Harmon did not get the committee chair, even though she was the senior member. Traditionally she would have, but the Speaker can override tradition. Harmon must have known it was probable, and was looking for an outside push in her favor.
Her trading a favor for a favor isn't treason, and I doubt it is even a real crime of any kind. She never offered to make a vote or even introduce legislation or hire anyone. She just said she'd do her job to help a particular person, in exchange for someone helping her. That is known as politics. Unfortunately, she was doing politics with an unsavory group in an unsavory way. And it was made public. Very Bad Luck.
For my next trick... Sausage Making !
No, we want to help these guys get the fine paid off. You do realize that not paying for "damages" is terms for being put back into prison, right? The property they "stole" was imaginary, but the money they have to pay has to be real.
The money in an electronic funds transfer is no more real than the "imaginary" property you refer to. So why is one any harder to deal with than the other? Its all just bits and bytes. Ignoring one should be of no more consequence that ignoring the other, right?
Get your arguments straight or step out of it.
These guys are heroes. They're putting their asses on the line for our right to copy - may it be equal to everyone else's - end copyright now.
Yes, part of this is about your right to copy. Personal copies, and fair use a part of this debate. But don't kid yourself. The serious issue here is not the right to copy. It is the right to copy and then distribute copies in a way that infringes on or negates the right of the creator to distribute or sell the product. In effect it is about the right to steal.
If you were distributing bread out of the back of a bakery without the baker's permission, on the theory that everyone has a right to bread, you might have plenty of takers. But you would have not a moral leg to stand on. It would not matter that you have a cool technological way to take the bread off the shelves, nor that the baker charges more that you think the bread should cost, nor that the bread would have gone uneaten had you not taken it and given it away. Nor would any other argument that gives you the power to distribute the bread in any way or for any reason that is against the wishes of its creator.
People would simply recognize it for the theft that it is. And that would be true, any way you slice it.
This entire story stinks of a distinct lack of personal responsiblity. As far as analogies go, think of it as someone who abandons a property for months on end, allowing the grass to grow high, paint to begin peeling off the siding, and animals to take up residence in the living room. The owner returns to said derelict property and is shocked to find a family of raccoons nesting in his lounger.
You insensitve clod! My home only looks that way to match the neighbors!
You think it is a coincidence your coworkers usually work in dank and dimly lit cubicles with no human contact? That they startle whenever anyone approaches? That they always bitch about the poor conditions at the rare and always uncomfortably awkward staff meetings, but no one ever actually attempts to change it? You're seeing it *EVERY DAY*
The whole point of a good snare is to put it in place without being noticed or felt. I'd like to think Time Warner will be very reasonable and rational about placing the limits. Put them where only 1, or 2, or 10% of the customers ever exceed them. 3+ standard deviations above the mean, etc. Something easily rationalized to a technically minded consumer.
But then the bean counters and MBAs will roll in. No matter where the limits are placed, they will try to find a way to squeeze out more profit and provide less service. Profit optimization. More and more people will find they have somehow tripped one of the new magic "penalty" clauses that will pop up every month. As more and more video and entertainment apps come along, its guaranteed that downloads and download volume will increase over time. Suddenly, you are an "excessive" user, and the jacked up rates apply.
The cost to (to the company) do so will decrease, but the price to the consumer will be driven by corporate desire for profit. And seeing as cable company internet service is nearly a monopoly, what are the chances they will drop prices as their own costs drop?
CADIE AI was base ball mad. ...
Had the fever and had it bad;
Just to root for the home town crew,
Ev'ry sou CADIE blew.
On a Saturday, her young beau
Called to (see (if (she'd like to go))),
To (see (a show)) but Miss CADIE said,
"No, I'll tell you what you can do."
Nahh. NPR is biased liberal to a large degree, but mostly in the features coverage. The news is pretty much straight off the wires field reporting.
If you want REAL left wing wild eyed stuff, listen to Pacifica radio news. They have core stations in New York, Washington, LA, and Berkeley. Just like Fox, but hitting you from the "left progressive" stance.
I can't watch Fox at home because I'd end up breaking my new plasma screen. But I can listen to Pacifica in the car because I can't really hit the radio buried in the dash.
Reporter bias, editor bias, assignment bias, publisher bias. Take your choice.
The only people not biased are people that don't have any background in the subject. And they are worthless too, since they don't understand what they are seeing and hearing.
Which leads us to how Entertainment Tonight covers political campaigns. Or maybe thats ABCNNBCBS Faux sensati-celebri-news. Hard to tell anymore.
We use Outlook/Exchange. It had a message recall button, but the function wasn't enabled. Which meant about once a week you could see a message with a followup THAT ASKED THE READER if the previous message could be recalled. Even if you said yes (after reading the mistaken message of course) the bad message did not disappear.
This worked great a flag for screwed up mail to be read first. Thanks Microsoft!
2:1 is probably only just noticeable, assuming it isn't an actual timed test. Anything that highly depends on user responsiveness (i.e. gaming and simulations) needs pretty dramatic pickups before the user will categorically agree it is better.
Even the more reasonable ones will want a "meaningful" increase. So the time saved has to be enough that they could do something useful with it. i.e. shoot another bad guy, beat the market to a good deal, go out for a smoke break, get home an hour earlier, etc.