In fact, historically speaking, the English word "American" was used to mean inhabitants and features of the English colonies in North America, which later evolved into the United States of America. The part that became Canada was not under British control until after the Battle of Quebec (1757, I think), and even after that, it was quite usual to use "America" or "Americans" in its by-then historically-established sense as referring to the thirteen older British colonies, and to use the term "Canada" to refer to the recently-conquered formerly-French colonies.
Being a natural language issue, it's not perfect, since "America" also included former Dutch possessions, so you don't get a nice rule out of it, like "America == British-established colonies", since not all of the 13 American colonies were of British origin.
But, the point is, an educated British subject of, say, 1776, would have thought it odd to use the term "Americans" to refer even to English inhabitants of Canada.
Of course, it's a source of irritation to folks who think that natural language usage ought to flow from axioms, that same word which describes the American continent does not, historically speaking, also describe all of its inhabitants, but the simple fact is that, for most language users, usage flows from history rather than axioms, and so obvious inconsistencies persist.
Note also that the scope of this discussion is restricted to the *English* words "America" and "Americans". I have no expertise in other languages (I am an American, after all...), and don't pretend to know their preferences on this.
Indeed, but if you google "OS X Lion boot disk" (not quoted, although that would probably work too), you'll find that the same instructions are all over the place. Basically, download the Lion installer, dig around inside it for a.dmg file, burn that, and you're done.
Microsoft (and many others) use copyright law to control my right to run their software, because, they assert, in order to get the software from the distribution medium into my computer where it can be run, I must make a copy of the software, and the legal mechanism whereby they exert their control is by specifying, by their copyright, the conditions under which I may make this copy.
I assert, on the same basis, that the nature of digital media are such that any technically feasible eavesdropping process involves making a copy of some portion of the stream, and transforming that copy into human-listenable or human-readable form. There is nothing in the law that says that the obligations of the recording-maker are reduced if the lifetime of the recording is brief, or if their intent is somehow benign. Just as a copy is a copy, a recording is a recording, and so any "digital eavesdropper" must, by virtue of the recording made as part of the eavesdropping process, obey the restrictions and obligations incumbent on any other overt or covert recording-maker.
I'm dreaming, obviously, but it would be nice to see that particular legal knife cut both ways, or none.
Re:This is seriously a world first?!!??
on
USB Foot Controls
·
· Score: 4, Funny
> I always wondered why they never took off.
Because they're on the floor, stuck under your foot?
So phones are maybe a reasonable proxy for riders, but it seems to me that the MTA, being a closed system which already has entry controls for fare-collection purposes, just might have more direct ways of getting at ridership, including ones that don't have smartphones. They could instrument their exit gates, too, and correlate them with train arrival times, and figure out which trains are letting off lots of passengers at which stations.
But maybe the phone thing still makes sense if they're not anonymizing it.
This is called "post-blogging" by a few bloggers, who do it. The most prominent one I've seen is the NYTimes "Opinionator" blogs, post-blogging the US civil war, to coincide with the 150th anniversary. I've also seen some history blogs do some more specific ones -- the Sudeten Crisis and the Blitz, in one case, and the Boxer rebellion in another.
I am ashamed that I had to look that up, but I'm not sure I like the analogy, assuming it's essentially NNTP with DVD-sized "articles". If users have to peer the whole feed (rather than just their subscriptions), then you'd need a lot more cache than 4GB, so maybe ISPs would run the peering servers? And in the video distribution model, individual users don't add new DVDs the way we used to add new articles to newsfeeds, the original data still comes from a central source, so maybe the peers are just forward caches, and aren't they already doing that with Akamai-style co-located equipment at all the big ISPs?
I've pretty much always been in favor of ATMs, but that's because I'm relatively anti-social -- I certainly recall the hue and cry about how impersonal and awful it was when they first became common. (Yes, I'm old, get off my lawn, etc.)
I'm still cranky about ATM fees, though -- the other thing I recall from when they were introduced was how much money the banks would save by not having to hire as many tellers, and these savings would more than cover the cost of the machines, so of course there would never be fees, they said. Simple common sense.
I'm just getting into KDE4 myself, seeing if I want to stick with it after my Debian "squeeze" upgrade, and so far, so good, mostly it's fine, just a bit unfamiliar. That's my take on it, anyways.
But, when I google for KDE4 config tips, I see pages and pages of rants about how it's is nothing but an Aero-esque fake Windows-inspired eye candy that doesn't even work, and how Ubuntu is way better because they have nice defaults and how these folks are switching to Ghome never ever ever coming back, etc. etc.
Pllus ca change, I suppose, but it's nice to see KDE getting some love.
> Trains can carry an entire ton of cargo 500 miles while expending a single gallon of diesel fuel.
While it's true that trains are efficient users of energy, I find this particular assertion especially maddening. I know the CSX publicity people like it, but it's not really true. What is true is that, once they are up to speed on level ground, a train can carry N tons of cargo 500 miles while expending G gallons of diesel fuel, and that N and G are approximately equal when N is around a few thousand
IMHO, this is not a trivial point about ratios, because the fact that the ratio of G to N is scale-dependent and only approaches one at large scales is very, very important when thinking about what rail does well versus what it does poorly.
Unit trains made up of hundreds of fully-loaded coal cars wending their way across America at 50 mph from mines in Wyoming to power plants in Texas, this is an excellent use of rail, and is very, very efficient. Likewise, moving grain from the Canaidan prairie to the seaports for export, and in general, moving mind-boggling quantities of buik commodities around at moderate speeds between a small number of widely-separated points is an excellent use of rail.
But if you have a lot of little customers all over the place, then it's not so good -- each customer needs a rail line, but it won't get much use, because each customer is small. Imagine trying to do FedEx or UPS with trains, and you'll get an idea of how rail's infrastructure requirements work against you, and how much of an energy advantage you'd need to make it worthwhile.
It is for this reason that, even as fuel prices increase, North American railways are abandoning their low-traffic spur lines and concentrating on the backbone part of their network, and investing in intermodal terminals so that they can containerized cargo on to trucks for the proverbial last mile.
It's also mostly for this reason that passenger rail tends to require subsidies. All that starting and stopping, and the time-critical nature of the "cargo", erode rail's energy advantage.
High speed rail is a whole other animal, it's purpose-built, with special locomotives and cars, and often a right-of-way that's not part of the cargo rail network at all. You gain a lot, but you lose the energy advantage pretty quickly.
Mandatory nit-pick: Turbines are also internal combustion engines.
Also, it's been done, although at a larger scale -- the Union Pacific Railroad used to have these beauties, which they liked for exactly the flex-fuel reason you mention. They could run on ultra-cheap heavy diesel oil.
They weren't especially efficient users of energy, however -- turbines are great at full power, but at low power loads, they still use almost as much fuel per unit time as at full power.
In a modern hybrid application, where there's a large battery to store energy, it may well work better.
> So... I hold this: that all of America is on a quaint, but working, well developed, almost universally understood system of measures, is no problem at all: when I want a 2 by 4, I don't want a 25mm by 50mm. Two... by... four.
I trust you are aware that a standard American two-by-four wooden board is actually 1.5 x 3.5 inches? And that for "dimensional lumber" (term i just learned from Wikipedia), only the specified length actually corresponds to the exact measure?
Of course, people die crossing the street, too. Maybe even Tyrone, whose replacement turns out to be worse.
On a good day, you can manage risk, but even on your best day, you can't eliminate it.
One apparent pebble-bed failure mode, just from looking at the Wikipedia page, is a moderator fire. Construct your own scenario, but if you get oxygen in that gas-cooling circuit at high enough temperatures, it's not hard to imagine an end game that features significant quantities of radioactive smoke and ash being vented to the local atmosphere.
> If Sony didn't ban cracked PS3's from PSN, and my gaming experience was affected by active exploits...
So here's a radical thought -- how about instead of curtailing the home-brew experience because of the fear of exploits and piracy, Sony actually allow modified consoles on the network, and not actually punish anyone until there's actual evidence of an exploit or piracy? This way they can respect people's rights and keep the network safe.
Or is respecting their customers just too much work?
Changing magnetic deviation due to movement of the magnetic pole goes on all the time. Runways are numbered according to their magnetic heading, plus or minus five degrees, and they have to keep them up to date, is all.
Two seconds of googling found this comment thread discussing a different runway-renumbering from July of 2009.
Obviously not enough airplane geeks around here...
Actually, it has been tried, though with magazine websites and not apps.
"Slate" magazine used to be a for-pay site, it cost around $20 per year to join, and you could get any/all content on the site, with (initially) no ads. Non-subcribers could not get any site content.
The problem, as it was described at the time, was that web surfing is not an orderly traversal of web server directories. People like to read an article, then surf away to another site for a while, maybe to view related material or whatever, and then come back. Their trajectory will traverse numerous websites and servers, and pay-walls don't make people subscribe, they make them leave. Consequently, the pay-walled magazine has a hard time using the web to promote itself -- it can't weave itself into viewers' trajectories, all it can do is advertise on other, often competing, sites. Slate went to a free, ad-supported mode after about two years as a pay service. (I was a charter subscriber, so I have demonstrated a willingness to pay -- didn't do me a lick of good.)
Salon has a similar issue, their solution is their famous annoy-o-tron, they'll show you the lede, and then hit you with an add, remarking tersely that subscribers don't have to see the ad. Or at least, they used to. I used to have a Salon subscription too, but it lapsed, and I found I didn't really miss it -- that one's my fault.
I dimly recall (but can't find the reference) that Tim O'Reilly (of the famous technical publisher) had a good article a few years ago on the essential dynamic that drives this process -- the fundamental idea is that, by lowering barriers to access and recouping costs by hosting ads, you broaden your potential audience enormously (orders of magnitude), because readers don't have to be choosy, they can sample. People get choosy when they have to pay. How would the web economics work if every web site cost ten cents to visit, but you only had to pay it once in your life per website? Would you fork over a dime for every new site? Would the sites make enough money to operate without ads? What about five cents? What about one? Practical experience suggests that the turn-over point in this analysis, for most websites, is so close to zero cents that it's not worth the infrastructural burden to implement it.
Also, my counter-example of failures included demonstrated history from the pay-TV world, which started out ad-free, but which was unstable, and soon got ads. Because they got greedy? Because they actually couldn't support themselves? I don't know, but I do know that the did try it, and it didn't last.
Hell, for that matter, print magazines have ads in 'em. Even high-toned niche magazines.
I think the only ad-free content I ever see these days, pay or otherwise, is printed books and some (but not all) scientific journals. The fact that absolutely every other medium, pay or otherwise, has ads in it, strongly suggests to me that the reason for this is something deeper than "they did not try."
That's a nice idea, but it's not going to work, they've already gone in a different direction. Some magazines offer bits of supplementary web-site content to their print subscribers. It's a real premium, but it doesn't get you away from the ads.
Also, it's happened before, your TBS example is instructive -- we used to think that subscription TV was supposed to be ad-free, but that didn't last very long.
SCTV had a good Dave Thomas bit about it, which is inexplicably not on YouTube...
> Why? Do rear-view window brakelights alert the drivers behind you better...
Yes, they do. In particular, the so-called "cyclops" does not come on with the headlights, only with the brakes, with the result that "car ahead has lights on" and "car ahead is braking" give different configurations of lights, not just different brightnesses. The change in configuration is more attention-grabbing than just brightening an already-existing light configuration.
I am all in favor of a new and better graphical system, but for the love of God, PLEASE keep network transparency. I want to forward my graphical session to other hosts, and have windows from remote systems show up on (and be managed by) my local display. This is *essential* for some sysadmin tasks I have to do, on a remote system that *has* *no* *graphical* *console*, but for which some of the tools *require* a GUI. At the moment, the saving grace of this system is system is that I can ssh in, forward my X connection, and run GUI software remotely.
On a related note, I wish to inform the community at large and Ubuntu in particular that not everyone is using a personally-administered workstation with a local file system. Some of us NFS-mount our home directories from a central server, and some of us install software on application servers which are also NFS-mounted. Please take care that "new improved" installers and desktop systems do not break in this environment.
Isn't one of the advantages of Android the ability to install apps from other than the Google app store? So people who really want this thing can still get it, independently of Google's disapproving glare, right?
Slashdot Mirror
In fact, historically speaking, the English word "American" was used to mean inhabitants and features of the English colonies in North America, which later evolved into the United States of America. The part that became Canada was not under British control until after the Battle of Quebec (1757, I think), and even after that, it was quite usual to use "America" or "Americans" in its by-then historically-established sense as referring to the thirteen older British colonies, and to use the term "Canada" to refer to the recently-conquered formerly-French colonies.
Being a natural language issue, it's not perfect, since "America" also included former Dutch possessions, so you don't get a nice rule out of it, like "America == British-established colonies", since not all of the 13 American colonies were of British origin.
But, the point is, an educated British subject of, say, 1776, would have thought it odd to use the term "Americans" to refer even to English inhabitants of Canada.
Of course, it's a source of irritation to folks who think that natural language usage ought to flow from axioms, that same word which describes the American continent does not, historically speaking, also describe all of its inhabitants, but the simple fact is that, for most language users, usage flows from history rather than axioms, and so obvious inconsistencies persist.
Note also that the scope of this discussion is restricted to the *English* words "America" and "Americans". I have no expertise in other languages (I am an American, after all...), and don't pretend to know their preferences on this.
Indeed, but if you google "OS X Lion boot disk" (not quoted, although that would probably work too), you'll find that the same instructions are all over the place. Basically, download the Lion installer, dig around inside it for a .dmg file, burn that, and you're done.
Well, turnabout being fair play, how's this:
Microsoft (and many others) use copyright law to control my right to run their software, because, they assert, in order to get the software from the distribution medium into my computer where it can be run, I must make a copy of the software, and the legal mechanism whereby they exert their control is by specifying, by their copyright, the conditions under which I may make this copy.
I assert, on the same basis, that the nature of digital media are such that any technically feasible eavesdropping process involves making a copy of some portion of the stream, and transforming that copy into human-listenable or human-readable form. There is nothing in the law that says that the obligations of the recording-maker are reduced if the lifetime of the recording is brief, or if their intent is somehow benign. Just as a copy is a copy, a recording is a recording, and so any "digital eavesdropper" must, by virtue of the recording made as part of the eavesdropping process, obey the restrictions and obligations incumbent on any other overt or covert recording-maker.
I'm dreaming, obviously, but it would be nice to see that particular legal knife cut both ways, or none.
> I always wondered why they never took off.
Because they're on the floor, stuck under your foot?
Duh.
So phones are maybe a reasonable proxy for riders, but it seems to me that the MTA, being a closed system which already has entry controls for fare-collection purposes, just might have more direct ways of getting at ridership, including ones that don't have smartphones. They could instrument their exit gates, too, and correlate them with train arrival times, and figure out which trains are letting off lots of passengers at which stations.
But maybe the phone thing still makes sense if they're not anonymizing it.
This is called "post-blogging" by a few bloggers, who do it. The most prominent one I've seen is the NYTimes "Opinionator" blogs, post-blogging the US civil war, to coincide with the 150th anniversary. I've also seen some history blogs do some more specific ones -- the Sudeten Crisis and the Blitz, in one case, and the Boxer rebellion in another.
> The custom port was 119.
I am ashamed that I had to look that up, but I'm not sure I like the analogy, assuming it's essentially NNTP with DVD-sized "articles". If users have to peer the whole feed (rather than just their subscriptions), then you'd need a lot more cache than 4GB, so maybe ISPs would run the peering servers? And in the video distribution model, individual users don't add new DVDs the way we used to add new articles to newsfeeds, the original data still comes from a central source, so maybe the peers are just forward caches, and aren't they already doing that with Akamai-style co-located equipment at all the big ISPs?
Or maybe I'm just missing something...
I've pretty much always been in favor of ATMs, but that's because I'm relatively anti-social -- I certainly recall the hue and cry about how impersonal and awful it was when they first became common. (Yes, I'm old, get off my lawn, etc.)
I'm still cranky about ATM fees, though -- the other thing I recall from when they were introduced was how much money the banks would save by not having to hire as many tellers, and these savings would more than cover the cost of the machines, so of course there would never be fees, they said. Simple common sense.
> ...or the Prime Minister of the UK as "Gordon"...
Particularly odd, as the Prime Minister of the UK has been David Cameron for a year or so...
Wow.
I'm just getting into KDE4 myself, seeing if I want to stick with it after my Debian "squeeze" upgrade, and so far, so good, mostly it's fine, just a bit unfamiliar. That's my take on it, anyways.
But, when I google for KDE4 config tips, I see pages and pages of rants about how it's is nothing but an Aero-esque fake Windows-inspired eye candy that doesn't even work, and how Ubuntu is way better because they have nice defaults and how these folks are switching to Ghome never ever ever coming back, etc. etc.
Pllus ca change, I suppose, but it's nice to see KDE getting some love.
> Trains can carry an entire ton of cargo 500 miles while expending a single gallon of diesel fuel.
While it's true that trains are efficient users of energy, I find this particular assertion especially maddening. I know the CSX publicity people like it, but it's not really true. What is true is that, once they are up to speed on level ground, a train can carry N tons of cargo 500 miles while expending G gallons of diesel fuel, and that N and G are approximately equal when N is around a few thousand
IMHO, this is not a trivial point about ratios, because the fact that the ratio of G to N is scale-dependent and only approaches one at large scales is very, very important when thinking about what rail does well versus what it does poorly.
Unit trains made up of hundreds of fully-loaded coal cars wending their way across America at 50 mph from mines in Wyoming to power plants in Texas, this is an excellent use of rail, and is very, very efficient. Likewise, moving grain from the Canaidan prairie to the seaports for export, and in general, moving mind-boggling quantities of buik commodities around at moderate speeds between a small number of widely-separated points is an excellent use of rail.
But if you have a lot of little customers all over the place, then it's not so good -- each customer needs a rail line, but it won't get much use, because each customer is small. Imagine trying to do FedEx or UPS with trains, and you'll get an idea of how rail's infrastructure requirements work against you, and how much of an energy advantage you'd need to make it worthwhile.
It is for this reason that, even as fuel prices increase, North American railways are abandoning their low-traffic spur lines and concentrating on the backbone part of their network, and investing in intermodal terminals so that they can containerized cargo on to trucks for the proverbial last mile.
It's also mostly for this reason that passenger rail tends to require subsidies. All that starting and stopping, and the time-critical nature of the "cargo", erode rail's energy advantage.
High speed rail is a whole other animal, it's purpose-built, with special locomotives and cars, and often a right-of-way that's not part of the cargo rail network at all. You gain a lot, but you lose the energy advantage pretty quickly.
Mandatory nit-pick: Turbines are also internal combustion engines.
Also, it's been done, although at a larger scale -- the Union Pacific Railroad used to have these beauties, which they liked for exactly the flex-fuel reason you mention. They could run on ultra-cheap heavy diesel oil.
They weren't especially efficient users of energy, however -- turbines are great at full power, but at low power loads, they still use almost as much fuel per unit time as at full power.
In a modern hybrid application, where there's a large battery to store energy, it may well work better.
> So... I hold this: that all of America is on a quaint, but working, well developed, almost universally understood system of measures, is no problem at all: when I want a 2 by 4, I don't want a 25mm by 50mm. Two... by... four.
I trust you are aware that a standard American two-by-four wooden board is actually 1.5 x 3.5 inches? And that for "dimensional lumber" (term i just learned from Wikipedia), only the specified length actually corresponds to the exact measure?
Nice.
Of course, people die crossing the street, too. Maybe even Tyrone, whose replacement turns out to be worse.
On a good day, you can manage risk, but even on your best day, you can't eliminate it.
One apparent pebble-bed failure mode, just from looking at the Wikipedia page, is a moderator fire. Construct your own scenario, but if you get oxygen in that gas-cooling circuit at high enough temperatures, it's not hard to imagine an end game that features significant quantities of radioactive smoke and ash being vented to the local atmosphere.
> If Sony didn't ban cracked PS3's from PSN, and my gaming experience was affected by active exploits...
So here's a radical thought -- how about instead of curtailing the home-brew experience because of the fear of exploits and piracy, Sony actually allow modified consoles on the network, and not actually punish anyone until there's actual evidence of an exploit or piracy? This way they can respect people's rights and keep the network safe.
Or is respecting their customers just too much work?
Ah, yes, nntp, one of the greats.
Of course, if you've set up your phone to be an NNTP server, you're pretty much asking for it in terms of bandwidth consumption.
Also, I think phones get their TOD from the cellular network, don't they?
Surely you meant, "striking Eul".
HTH.
The choosing is weirding, you're right, but as long as the meaning is clearing, it's not really probleming.
Changing magnetic deviation due to movement of the magnetic pole goes on all the time. Runways are numbered according to their magnetic heading, plus or minus five degrees, and they have to keep them up to date, is all.
Two seconds of googling found this comment thread discussing a different runway-renumbering from July of 2009.
Obviously not enough airplane geeks around here...
Actually, it has been tried, though with magazine websites and not apps.
"Slate" magazine used to be a for-pay site, it cost around $20 per year to join, and you could get any/all content on the site, with (initially) no ads. Non-subcribers could not get any site content.
The problem, as it was described at the time, was that web surfing is not an orderly traversal of web server directories. People like to read an article, then surf away to another site for a while, maybe to view related material or whatever, and then come back. Their trajectory will traverse numerous websites and servers, and pay-walls don't make people subscribe, they make them leave. Consequently, the pay-walled magazine has a hard time using the web to promote itself -- it can't weave itself into viewers' trajectories, all it can do is advertise on other, often competing, sites. Slate went to a free, ad-supported mode after about two years as a pay service. (I was a charter subscriber, so I have demonstrated a willingness to pay -- didn't do me a lick of good.)
Salon has a similar issue, their solution is their famous annoy-o-tron, they'll show you the lede, and then hit you with an add, remarking tersely that subscribers don't have to see the ad. Or at least, they used to. I used to have a Salon subscription too, but it lapsed, and I found I didn't really miss it -- that one's my fault.
I dimly recall (but can't find the reference) that Tim O'Reilly (of the famous technical publisher) had a good article a few years ago on the essential dynamic that drives this process -- the fundamental idea is that, by lowering barriers to access and recouping costs by hosting ads, you broaden your potential audience enormously (orders of magnitude), because readers don't have to be choosy, they can sample. People get choosy when they have to pay. How would the web economics work if every web site cost ten cents to visit, but you only had to pay it once in your life per website? Would you fork over a dime for every new site? Would the sites make enough money to operate without ads? What about five cents? What about one? Practical experience suggests that the turn-over point in this analysis, for most websites, is so close to zero cents that it's not worth the infrastructural burden to implement it.
Also, my counter-example of failures included demonstrated history from the pay-TV world, which started out ad-free, but which was unstable, and soon got ads. Because they got greedy? Because they actually couldn't support themselves? I don't know, but I do know that the did try it, and it didn't last.
Hell, for that matter, print magazines have ads in 'em. Even high-toned niche magazines.
I think the only ad-free content I ever see these days, pay or otherwise, is printed books and some (but not all) scientific journals. The fact that absolutely every other medium, pay or otherwise, has ads in it, strongly suggests to me that the reason for this is something deeper than "they did not try."
That's a nice idea, but it's not going to work, they've already gone in a different direction. Some magazines offer bits of supplementary web-site content to their print subscribers. It's a real premium, but it doesn't get you away from the ads.
Also, it's happened before, your TBS example is instructive -- we used to think that subscription TV was supposed to be ad-free, but that didn't last very long.
SCTV had a good Dave Thomas bit about it, which is inexplicably not on YouTube...
> Why? Do rear-view window brakelights alert the drivers behind you better...
Yes, they do. In particular, the so-called "cyclops" does not come on with the headlights, only with the brakes, with the result that "car ahead has lights on" and "car ahead is braking" give different configurations of lights, not just different brightnesses. The change in configuration is more attention-grabbing than just brightening an already-existing light configuration.
So how long was this in upstream, and which distros have packaged up the broken one?
Debian's got 1.3.1 in stable, and 1.3.3 in squeeze, so the latter might be built from the compromised one.
Others?
I am all in favor of a new and better graphical system, but for the love of God, PLEASE keep network transparency. I want to forward my graphical session to other hosts, and have windows from remote systems show up on (and be managed by) my local display. This is *essential* for some sysadmin tasks I have to do, on a remote system that *has* *no* *graphical* *console*, but for which some of the tools *require* a GUI. At the moment, the saving grace of this system is system is that I can ssh in, forward my X connection, and run GUI software remotely.
On a related note, I wish to inform the community at large and Ubuntu in particular that not everyone is using a personally-administered workstation with a local file system. Some of us NFS-mount our home directories from a central server, and some of us install software on application servers which are also NFS-mounted. Please take care that "new improved" installers and desktop systems do not break in this environment.
Thank you.
Isn't one of the advantages of Android the ability to install apps from other than the Google app store? So people who really want this thing can still get it, independently of Google's disapproving glare, right?
Genuinely curious about this.