How is division by halves any more clumsy in metric than in fractional measurements?
If I'm measuring something that's 105 mm and I need to halve it, it's pretty easy to know straight away that it's 52.5 mm How is this any harder to measuring something that's 4 and 3/16 inches and having to find half of that?
Cooking is very easy in metric.
1 litre of water = 1000 ml = 1 kg 1 teaspoon = 5 ml 1 tablespoon = 20 ml
1 cup = 250 ml 4 cups = 1 litre 1/2 cup = 125 ml
Regarding the trusty ol' 2x4 lumber - it's not actually 2" x 4" anyway, they're more like 1 1/2" x 3 x 1/2" which works out to be 38 x 89 mm. I think the equivalent of a 2x4 here though is a bit bigger at 50 x 100 mm and then the stud framing is generally on 400 mm centres.
As for 2 inches less a sixteenth (which, being 1 and fifteen sixteenths, is pretty clumsy to say) would be 50 mm less 2 mm and that works out at 48 mm, or if you're saying it, forty-eight mills (yes, I know a mil is officially defined as 1/1000 of an inch, but if you never use inches, the shorthand way of saying one millimetre is one mill)
My stride, and most people's stride, is around 600 mm - it's pretty easy to work from there. I don't know why you need to count paces and multiply and then divide, once again, that's the point of the metric system, it's all decimal.
If I walk, say, 85 strides, it's easy to calculate 83 * 6 and then move the decimal point on place to the left to get the distance in metres. 60 * 8 = 480 3 * 6 = 18 480 + 18 = 498 50 m
Either that, or shorten your stride for the purpose of measuring things to 0.5m and multiply by two.
With your example, if you need to know the distance in yards or miles, it's then more complicated than just counting paces and multiplying by two.
If you're putting a window in a wall, and measuring things in metric units, floating point rounding errors are never going to be a problem - they're going to be orders of magnitude smaller than your margin of error. If you're using a calculator then, to avoid such errors even potentially being a problem, it's most likely working internally with BCD anyway, so there are no floating point rounding errors and no having to work with fractions, so you can easily convert between, say, centimetres and metres or millimetres by shifting the decimal point in your head.
Quick - what's 1/2" + 7/32" + 3/8" ? Now, add up 50 mm + 20 mm + 4 mm and convert to cm.
I use kilometres-per-hour and kilometres as a measurement of driving distance. I use litres for petrol and milk. milliliters for other liquid measurements (since it's on the bottles, and easy enough to read) millimetres for building things out of wood. 1 mm is a good tolerance in construction. C for temperature (because 37.0 isn't any worse than 98.6)
And when I do science, I use metric. No prob and it's all nice and consistent across the board
1 mm is roughly 1/32 Inches - so right off the bat there you've got have the accuracy that you asked for.
Furthermore, it is possible to have decimal millimetres, whilst most construction units are given in mm (eg, a kitchen cabinet is generally 600mm wide, light switches are usually 1500mm above the floor) if you need more accuracy (and you generally don't in construction) then you can specify additional accuracy using decimal places... That's the whole point of the metric system, that using decimal places becomes easy. 1 metre = 100 cm, 10 cm = 0.1 m = 100 mm etc...
If you need to put something in the middle of two points that are, say, 105 mm apart, it's pretty straightforward to specify that the distance is 52.5 mm Rulers will often have 0.5 mm markings (again, roughly, 1/64 inch) for the first 50 or 100 mm. If your measure doesn't have half mm markings, it's usually accurate enough to interpolate by eye and put the mark on your materials half-way between two mm marks on the measurement device.
Because it all comes down to the resolution of the picture. They want as accurate ans as detailed a picture as they can get, and using a mosaic filter over the sensor firstly fixes you to a single set of colour filters, as it has to be precisely aligned with the image sensor elements, and you have to then interpolate image data (ie, guess at filling in missing colour information)
With a b&w sensor, you have more freedom in swapping out filters over the sensor and you aren't interpolating picture information.
Get enterprise series drives, not consumer drives. One difference is the firmware is a lot more up-front about errors, rather than trying to hide them and carry on as if everything is OK. In a RAID, you're going to want to fail a drive as soon as it starts to play up, whereas the average consumer wants a drive that doesn't turn around and die at the first small error, where it can remap sectors and pretend that nothing happened. Part of the reason enterprise drives cost more, when they're often the same, or very similar, physical hardware is that the price includes the better warranty...
Setting up a CA and distributing certs to a large number of workstations takes longer than logging onto InstantSSL/StartSSL/GoDaddy/Symantec and generating a certificate that's got a chain of trust to a root CA. Plus, if you're then having devices connecting from outside of your organisation, or devices you don't correctly control, then their mail clients or browsers don't throw the big certificate warning up at them. It's that, more than anything else, the way Firefox and Chrome in particular, handle self-signed certificates that they haven't been explicitly told to trust, that makes certs from a registered CA worthwhile.
If you're a large organisation, you still don't have a large number N of web-facing servers that need real SSL certificates. You might have a huge number of internal servers, and then absolutely you'll have your own internal CA, but for internet-facing servers that have incoming SSL connections to them, $30 for a cert on a $5-10k Exchange box is a drop in the ocean.
Anyway, for the case of what this thread was originally about, which is Google being able to connect to your mail server over POP3 secured with SSL and retrieve email, having a proper SSL cert absolutely is better security than it blindly accepting self-signed certificates when downloading your POP email into your gmail mailbox.
No, it's perfectly reasonable to run your own CA, as an individual or an organization, distribute your CA cert to those using the service, and go merrily on your encrypted and authenticated way.
For $30 per year to get a real cert (or even less, a little googling will quickly product things like 80% off at GoDaddy etc), your time has to be of quite a low value if it's easier/cheaper to run your own CA and distribute certificates (unless, of course, you're doing it all for the fun of it)
Where self-signed certs are no good is when you need to access your SSL protected service from someone else's machine, or a machine you've not used to access the service from before, and you have to take it on blind faith (or remember a long and complicated fingerprint) that the cert you're getting is the correct one.
The paying to get a SSL certificate only affects people running a mail server, not people using a mail server. If you're running a mail server, you should really get a recognised SSL certificate if you want to offer SSL protected services, otherwise you're only getting half the benefit of SSL connections - you get encryption but not authentication.
From my reading of the linked article, this has nothing whatsoever to do with fetching your email from Google over POP3 (or POP3S)
What this affects is if you are running a mailserver that uses a self-signed certificate, or if you're using another email account on a mailserver that uses a self-signed certificate, then you can no longer tell your gmail account to pull the email in from your second account over POP3S, as it can't verify the certificate.
You can still have gmail pull in your POP email via the non-secure protocol, or have the mail server administrator pay the $30 or so a year it costs to get a valid certificate signed by a recognissed CA.
You can still fetch your gmail via POP, using SSL or not, although why anyone would want to use POP if they're given any other option (such as IMAP) is beyond me.
From my reading of the linked article, this has nothing whatsoever to do with fetching your email from Google over POP3 (or POP3S)
What this affects is if you are running a mailserver that uses a self-signed certificate, or if you're using another email account on a mailserver that uses a self-signed certificate, then you can no longer tell your gmail account to pull the email in from your second account over POP3S, as it can't verify the certificate.
You can still have gmail pull in your POP email via the non-secure protocol, or have the mail server administrator pay the $30 or so a year it costs to get a valid certificate signed by a recognissed CA.
You can still fetch your gmail via POP, using SSL or not, although why anyone would want to use POP if they're given any other option (such as IMAP) is beyond me.
It may seem less, but I have this thing against throwing money away when I don't have to....
Except that you don't. Time and time again, sales analysis has shown that iOS users are overwhelmingly more likely to pay for apps than Android users. If you're selling your app for $0.99 then you only need to sell 100 more copies for iOS than for Android before you've broken even...
The developer of DOSBox Turbo is refusing to release the source for his application unless you pay the $3.99 to "buy" a license of it. The same developer explicitly states that the "small" fee (although one might argue that $3.99 is pretty expensive for an OSS Android app)
This sounds exactly like he's complaining about having to pay $4 for the app, and is claiming that it's expensive.
There's nothing in the GPL that prevents you from selling GPL'd software.
The alleged GPL violation for DOSBot is another issue altogether, and if it were truly about GPL violations, then why even bring DOSBox Turbo into the discussion as they seem to be complying with the GPL by offering the source if you buy the app.
What kind of crazy world do we live in where someone can own a multiple hundred dollar smartphone made by a large, faceless corporation, pay for the plan that allows this device to connect to a network and send/receive calls and data and yet this person feels that a measly four bucks, the cost of a cup of coffee that they buy every day without giving it a second thought, four bucks is considered expensive for some software that took many, many man-hours for an individual person to develop (even if based on open source, the actual app didn't spring fully-formed from the source repository of DOSBox). Some software that allows them to relive their childhood, play some great games that they remember from the dawn of gaming and a measly four bucks is so expensive that it's a "small" fee?
If you get the paid version of Google Apps, you get a support ID number (or something along those lines) and provided with the details of how to contact Google. If you're on the free version, you don't get anything - although as you said, Google employees are often in the forums that you can use.
Having used the free version for a number of domains, I was under the impression that the free version had community support as it's only option - ie, you could post about your problems on the forums, but you couldn't raise an issue directly with Google.
Air freight is expensive, unless you pre-allocate so much capacity that the operators give you amazing rates. Apple are famous for one holiday season pre-purchasing so much additional air freight that PC manufacturers were left stranded because there were no more spots left on the planes.
Apple (historically) have used Singapore and then China as their manufacturing base and CTO machines are configured to your spec - quite literally, your order goes into a queue and the next machine off the assembly line is configured to your spec, and air-freighted to you. This is why a CTO will generally take two weeks, unless it's a common "custom" configuration that Apple have in stock.
The assembly is certainly not done like other vendors do, which is to pick parts from a bin and assemble your unit in the USA (or whatever country you're ordering from) they are configured in the same factory that builds the standard machines.
It's amazing the rates you can get for freight when you buy lots of capacity. I can order a 20kg server from my wholesaler in Sydney which is nearly 900km away. If my order is more than about $300, the freight is free. Otherwise it's $15. If I have to RMA this same 20kg server, it's going to cost me the best part of $100 to get it back up to Sydney.
Yep, although from the article I read, it looks like they time how long a frame takes to render. If that frame takes more than 1/60th second, then the next frame has it's horizontal resolution decreased by 32px and is then scaled to size. If that frame takes longer than 1/60th second, they repeat it again, knocking off 32 pixels of horizontal resolution each time.
As you say, it only happens when there's a lot of shit going on, and the blur is only in the horizontal direction, and because there's a lot going on, you don't notice it.
Can't say that I've ever noticed anything getting blurry in the horizontal direction when there's a lot happening, I'm always too busy trying to not crash or die.
They do also do some serious tricks with the SPUs, things like lighting and particle effects...
Now I may be wrong, but I believe that WipEout HD/Fury is proper 1080p60. That was one of the big things that held up the game - Studio Liverpool weren't going to ship it until it was running smoothly at Full HD.
From reading the interview, it seems they had to decide between 1080p and 720p with 2xMSAA and chose 1080p although it was a lot harder, they wanted to push the boundaries. They also implement a few cheats as well, things like dynamically altering the horizontal resolution (and then, I assume, scaling it up to 1920 pixels wide)
Yeah, I've not used AirVision, but have looked at it a bit of detail. The software needs some very basic improvements - even simple things like managing disk space and not just recording until the disk is full with no way to automatically clean up old recordings...
You don't want to go through the trouble and expense of rolling out cable to people's houses - you don't have the budget to cover for it, and no one could afford the installation charge if you passed it all on to them. Look at Ubiquity wireless gear - it's very good, priced amazingly well, and is relatively easy to set up and configure. They do backhaul stuff, distribution stuff and even 802.11a/b/g/n that is comparable to Cisco at 1/4 the price.
I'll second that - the easiest way to add a workgroup printer to a Windows machine is to install Bonjour print services for Windows. Discover the printer, hit done and it's all set up for you. Even if the printer is on a dynamic IP (as most are from the factory) it doesn't matter if it's IP changes as Zeroconf/Bonjour will find it with no worries. It's also good for auto-discovery of network services. Servers can advertise themselves over Bonjour, Safari can even display a Bonjour bookmark that shows printers and other web pages that are advertising themselves on the network. It really does make life easier...
Slashdot Mirror
How is division by halves any more clumsy in metric than in fractional measurements?
If I'm measuring something that's 105 mm and I need to halve it, it's pretty easy to know straight away that it's 52.5 mm
How is this any harder to measuring something that's 4 and 3/16 inches and having to find half of that?
Cooking is very easy in metric.
1 litre of water = 1000 ml = 1 kg
1 teaspoon = 5 ml
1 tablespoon = 20 ml
1 cup = 250 ml
4 cups = 1 litre
1/2 cup = 125 ml
Regarding the trusty ol' 2x4 lumber - it's not actually 2" x 4" anyway, they're more like 1 1/2" x 3 x 1/2" which works out to be 38 x 89 mm.
I think the equivalent of a 2x4 here though is a bit bigger at 50 x 100 mm and then the stud framing is generally on 400 mm centres.
As for 2 inches less a sixteenth (which, being 1 and fifteen sixteenths, is pretty clumsy to say) would be 50 mm less 2 mm and that works out at 48 mm, or if you're saying it, forty-eight mills (yes, I know a mil is officially defined as 1/1000 of an inch, but if you never use inches, the shorthand way of saying one millimetre is one mill)
My stride, and most people's stride, is around 600 mm - it's pretty easy to work from there. I don't know why you need to count paces and multiply and then divide, once again, that's the point of the metric system, it's all decimal.
If I walk, say, 85 strides, it's easy to calculate 83 * 6 and then move the decimal point on place to the left to get the distance in metres.
60 * 8 = 480
3 * 6 = 18
480 + 18 = 498 50 m
Either that, or shorten your stride for the purpose of measuring things to 0.5m and multiply by two.
With your example, if you need to know the distance in yards or miles, it's then more complicated than just counting paces and multiplying by two.
If you're putting a window in a wall, and measuring things in metric units, floating point rounding errors are never going to be a problem - they're going to be orders of magnitude smaller than your margin of error.
If you're using a calculator then, to avoid such errors even potentially being a problem, it's most likely working internally with BCD anyway, so there are no floating point rounding errors and no having to work with fractions, so you can easily convert between, say, centimetres and metres or millimetres by shifting the decimal point in your head.
Quick - what's 1/2" + 7/32" + 3/8" ?
Now, add up 50 mm + 20 mm + 4 mm and convert to cm.
Which was easier?
I use kilometres-per-hour and kilometres as a measurement of driving distance.
I use litres for petrol and milk.
milliliters for other liquid measurements (since it's on the bottles, and easy enough to read)
millimetres for building things out of wood. 1 mm is a good tolerance in construction.
C for temperature (because 37.0 isn't any worse than 98.6)
And when I do science, I use metric. No prob and it's all nice and consistent across the board
1 mm is roughly 1/32 Inches - so right off the bat there you've got have the accuracy that you asked for.
Furthermore, it is possible to have decimal millimetres, whilst most construction units are given in mm (eg, a kitchen cabinet is generally 600mm wide, light switches are usually 1500mm above the floor) if you need more accuracy (and you generally don't in construction) then you can specify additional accuracy using decimal places... That's the whole point of the metric system, that using decimal places becomes easy. 1 metre = 100 cm, 10 cm = 0.1 m = 100 mm etc...
If you need to put something in the middle of two points that are, say, 105 mm apart, it's pretty straightforward to specify that the distance is 52.5 mm
Rulers will often have 0.5 mm markings (again, roughly, 1/64 inch) for the first 50 or 100 mm. If your measure doesn't have half mm markings, it's usually accurate enough to interpolate by eye and put the mark on your materials half-way between two mm marks on the measurement device.
Because it all comes down to the resolution of the picture. They want as accurate ans as detailed a picture as they can get, and using a mosaic filter over the sensor firstly fixes you to a single set of colour filters, as it has to be precisely aligned with the image sensor elements, and you have to then interpolate image data (ie, guess at filling in missing colour information)
With a b&w sensor, you have more freedom in swapping out filters over the sensor and you aren't interpolating picture information.
Get enterprise series drives, not consumer drives. One difference is the firmware is a lot more up-front about errors, rather than trying to hide them and carry on as if everything is OK.
In a RAID, you're going to want to fail a drive as soon as it starts to play up, whereas the average consumer wants a drive that doesn't turn around and die at the first small error, where it can remap sectors and pretend that nothing happened.
Part of the reason enterprise drives cost more, when they're often the same, or very similar, physical hardware is that the price includes the better warranty...
Setting up a CA and distributing certs to a large number of workstations takes longer than logging onto InstantSSL/StartSSL/GoDaddy/Symantec and generating a certificate that's got a chain of trust to a root CA. Plus, if you're then having devices connecting from outside of your organisation, or devices you don't correctly control, then their mail clients or browsers don't throw the big certificate warning up at them.
It's that, more than anything else, the way Firefox and Chrome in particular, handle self-signed certificates that they haven't been explicitly told to trust, that makes certs from a registered CA worthwhile.
In light of the subject of this story, I'd expect that they'll stop that working sometime in the near future...
If you're a large organisation, you still don't have a large number N of web-facing servers that need real SSL certificates. You might have a huge number of internal servers, and then absolutely you'll have your own internal CA, but for internet-facing servers that have incoming SSL connections to them, $30 for a cert on a $5-10k Exchange box is a drop in the ocean.
Anyway, for the case of what this thread was originally about, which is Google being able to connect to your mail server over POP3 secured with SSL and retrieve email, having a proper SSL cert absolutely is better security than it blindly accepting self-signed certificates when downloading your POP email into your gmail mailbox.
No, it's perfectly reasonable to run your own CA, as an individual or an organization, distribute your CA cert to those using the service, and go merrily on your encrypted and authenticated way.
For $30 per year to get a real cert (or even less, a little googling will quickly product things like 80% off at GoDaddy etc), your time has to be of quite a low value if it's easier/cheaper to run your own CA and distribute certificates (unless, of course, you're doing it all for the fun of it)
Where self-signed certs are no good is when you need to access your SSL protected service from someone else's machine, or a machine you've not used to access the service from before, and you have to take it on blind faith (or remember a long and complicated fingerprint) that the cert you're getting is the correct one.
The paying to get a SSL certificate only affects people running a mail server, not people using a mail server.
If you're running a mail server, you should really get a recognised SSL certificate if you want to offer SSL protected services, otherwise you're only getting half the benefit of SSL connections - you get encryption but not authentication.
From my reading of the linked article, this has nothing whatsoever to do with fetching your email from Google over POP3 (or POP3S)
What this affects is if you are running a mailserver that uses a self-signed certificate, or if you're using another email account on a mailserver that uses a self-signed certificate, then you can no longer tell your gmail account to pull the email in from your second account over POP3S, as it can't verify the certificate.
You can still have gmail pull in your POP email via the non-secure protocol, or have the mail server administrator pay the $30 or so a year it costs to get a valid certificate signed by a recognissed CA.
You can still fetch your gmail via POP, using SSL or not, although why anyone would want to use POP if they're given any other option (such as IMAP) is beyond me.
From my reading of the linked article, this has nothing whatsoever to do with fetching your email from Google over POP3 (or POP3S)
What this affects is if you are running a mailserver that uses a self-signed certificate, or if you're using another email account on a mailserver that uses a self-signed certificate, then you can no longer tell your gmail account to pull the email in from your second account over POP3S, as it can't verify the certificate.
You can still have gmail pull in your POP email via the non-secure protocol, or have the mail server administrator pay the $30 or so a year it costs to get a valid certificate signed by a recognissed CA.
You can still fetch your gmail via POP, using SSL or not, although why anyone would want to use POP if they're given any other option (such as IMAP) is beyond me.
It may seem less, but I have this thing against throwing money away when I don't have to....
Except that you don't. Time and time again, sales analysis has shown that iOS users are overwhelmingly more likely to pay for apps than Android users. If you're selling your app for $0.99 then you only need to sell 100 more copies for iOS than for Android before you've broken even...
Here's just one example:
http://www.guardian.co.uk/technology/appsblog/2012/dec/04/ios-android-revenues-downloads-country
The developer of DOSBox Turbo is refusing to release the source for his application unless you pay the $3.99 to "buy" a license of it. The same developer explicitly states that the "small" fee (although one might argue that $3.99 is pretty expensive for an OSS Android app)
This sounds exactly like he's complaining about having to pay $4 for the app, and is claiming that it's expensive.
There's nothing in the GPL that prevents you from selling GPL'd software.
The alleged GPL violation for DOSBot is another issue altogether, and if it were truly about GPL violations, then why even bring DOSBox Turbo into the discussion as they seem to be complying with the GPL by offering the source if you buy the app.
What kind of crazy world do we live in where someone can own a multiple hundred dollar smartphone made by a large, faceless corporation, pay for the plan that allows this device to connect to a network and send/receive calls and data and yet this person feels that a measly four bucks, the cost of a cup of coffee that they buy every day without giving it a second thought, four bucks is considered expensive for some software that took many, many man-hours for an individual person to develop (even if based on open source, the actual app didn't spring fully-formed from the source repository of DOSBox). Some software that allows them to relive their childhood, play some great games that they remember from the dawn of gaming and a measly four bucks is so expensive that it's a "small" fee?
Seriously, get a grip!
If you get the paid version of Google Apps, you get a support ID number (or something along those lines) and provided with the details of how to contact Google. If you're on the free version, you don't get anything - although as you said, Google employees are often in the forums that you can use.
Having used the free version for a number of domains, I was under the impression that the free version had community support as it's only option - ie, you could post about your problems on the forums, but you couldn't raise an issue directly with Google.
Interesting they didn't go all in with Apple and put it up on iTunes.
Air freight is expensive, unless you pre-allocate so much capacity that the operators give you amazing rates.
Apple are famous for one holiday season pre-purchasing so much additional air freight that PC manufacturers were left stranded because there were no more spots left on the planes.
Apple (historically) have used Singapore and then China as their manufacturing base and CTO machines are configured to your spec - quite literally, your order goes into a queue and the next machine off the assembly line is configured to your spec, and air-freighted to you. This is why a CTO will generally take two weeks, unless it's a common "custom" configuration that Apple have in stock.
The assembly is certainly not done like other vendors do, which is to pick parts from a bin and assemble your unit in the USA (or whatever country you're ordering from) they are configured in the same factory that builds the standard machines.
It's amazing the rates you can get for freight when you buy lots of capacity.
I can order a 20kg server from my wholesaler in Sydney which is nearly 900km away. If my order is more than about $300, the freight is free. Otherwise it's $15. If I have to RMA this same 20kg server, it's going to cost me the best part of $100 to get it back up to Sydney.
Yep, although from the article I read, it looks like they time how long a frame takes to render. If that frame takes more than 1/60th second, then the next frame has it's horizontal resolution decreased by 32px and is then scaled to size. If that frame takes longer than 1/60th second, they repeat it again, knocking off 32 pixels of horizontal resolution each time.
As you say, it only happens when there's a lot of shit going on, and the blur is only in the horizontal direction, and because there's a lot going on, you don't notice it.
Can't say that I've ever noticed anything getting blurry in the horizontal direction when there's a lot happening, I'm always too busy trying to not crash or die.
They do also do some serious tricks with the SPUs, things like lighting and particle effects...
Now I may be wrong, but I believe that WipEout HD/Fury is proper 1080p60.
That was one of the big things that held up the game - Studio Liverpool weren't going to ship it until it was running smoothly at Full HD.
http://www.eurogamer.net/articles/digitalfoundry-wipeout-hd-fury-interview
From reading the interview, it seems they had to decide between 1080p and 720p with 2xMSAA and chose 1080p although it was a lot harder, they wanted to push the boundaries. They also implement a few cheats as well, things like dynamically altering the horizontal resolution (and then, I assume, scaling it up to 1920 pixels wide)
Yeah, I've not used AirVision, but have looked at it a bit of detail. The software needs some very basic improvements - even simple things like managing disk space and not just recording until the disk is full with no way to automatically clean up old recordings...
You don't want to go through the trouble and expense of rolling out cable to people's houses - you don't have the budget to cover for it, and no one could afford the installation charge if you passed it all on to them. Look at Ubiquity wireless gear - it's very good, priced amazingly well, and is relatively easy to set up and configure. They do backhaul stuff, distribution stuff and even 802.11a/b/g/n that is comparable to Cisco at 1/4 the price.
I'll second that - the easiest way to add a workgroup printer to a Windows machine is to install Bonjour print services for Windows. Discover the printer, hit done and it's all set up for you. Even if the printer is on a dynamic IP (as most are from the factory) it doesn't matter if it's IP changes as Zeroconf/Bonjour will find it with no worries.
It's also good for auto-discovery of network services. Servers can advertise themselves over Bonjour, Safari can even display a Bonjour bookmark that shows printers and other web pages that are advertising themselves on the network. It really does make life easier...